Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Gaussian Process”

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 415 records · Page 23

Covariate Dependent Sparse Functional Data Analysis

This study proposes a method to incorporate covariate information into sparse functional data analysis. The method aims at cases where each subject has a limited number of longitudinal measurements and is associated with static covariates. This research is motivated by several use cases in practice. One representative example is void swelling, a nuclear-specific material degradation mechanism. Void swelling is affected by many covariates, including alloy composition and irradiation type. How to accurately model the complicated joint effects of such covariates on the swelling process is the key to mitigating the effect of swelling and ensuring safe operation. Unlike most of the existing methods, the proposed method can handle high-dimensional covariates with the informative covariate identification procedure and sparse and irregularly spaced measurements, that is, does not require complete or dense observations. The main innovation of the proposed method is that we model the variation coming from covariates and the variation left conditioned on covariates, such that the functional principal component analysis and Gaussian process can be conducted in a unified manner. Further, we also propose a systematic approach to identify important covariates in the hypothesis testing context. The methodology is demonstrated on applications in nuclear engineering and healthcare and simulation studies.

42 ENGINEERING↗

When physics-informed data analytics outperforms black-box machine learning: A case study in thickness control for additive manufacturing

Aerosol jet printing (AJP) has emerged as a promising noncontact additive manufacturing method for high-resolution printing for a wide range of material systems. A key challenge limiting the broader adoption of AJP in the material science community is the lack of methods to precisely control thickness. Herein, we develop a model-based design of experiment (MBDoE) framework that integrates physics-informed models, nonlinear regression, and information criteria to postulate, select and calibrate the best model to describe and optimize the AJP manufacturing process. Starting with already available data from system commissioning (e.g., prior single variable sensitivity analysis), four candidate physics-informed models are postulated and trained. MBDoE identifies a single additional optimal experiment to validate these predictive models with quantified uncertainties, which are then used to determine the best experimental conditions to control printed film thickness. As a comparative benchmark, the analysis is repeated using the same dataset with nonparametric Gaussian process regression (GPR) model that does not incorporate physical information. Using MBDoE principles, we find that only five experiments are necessary to calibrate the nonlinear physics-informed parametric model, and with said limited data, this model outperforms the black-box machine learning GPR model. This key result underscores an emerging trend in the data science community: incorporating physical information into predictive models often drastically reduces the data requirements. Leveraging MBDoE further increased the data efficiency. By design, the proposed data science framework is general in nature and can be easily extended to other experimental and additive manufacturing systems beyond AJP.

Aerosol jet printing↗

Big Data and AI at DoE's Legacy Sites - 20546

More than 30 years have passed since DOE started the decommissioning of nuclear weapon complexes and the clean-up of soil and groundwater. All the sites have been collecting and archiving soil and groundwater monitoring datasets; particularly contaminant concentration time-series. These datasets provide unparalleled opportunities to understand the system behavior (including more fundamental hydrological and geochemical processes, the response to various perturbations, the long-term trend and environmental decay rate towards the regulatory limit). This understanding is critical for providing multiple lines of evidences that can support site closure. In this study, we explore the machine learning (ML) and artificial intelligence (AI) applications to the long-term soil and groundwater management at DoE's legacy sites. ML can improve our understanding of the subsurface systems, which is critical for long-term monitoring and management of the sites, while AI can automate or support some of decision-making processes (e.g., anomaly detection, monitoring well placements). The particular focuses are to develop general algorithms to: (1) to identify distinct spatiotemporal patterns and to identify several groups that have similar temporal behaviors, using unsupervised clustering methods, (2) identify the different temporal scales of hydrological responses to climate perturbations by time-series analysis, and (3) reduce the number of monitoring wells by identifying the minimum sufficient number of wells to capture the heterogeneity of the groundwater contaminant plume and concentration distribution, using the Gaussian Process model. We demonstrate our methodology at the Savannah River Site F-Area. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Constrained Bayesian Optimization of Criticality Experiments at LLNL [Slides]

The design of criticality experiments is typically an iterative process that employs a Monte Carlo transport code. The goal is to find a design that optimizes some variable, like the sensitivity of a response to a cross section, while simultaneously ensuring criticality. The high fidelity of the Monte Carlo code is a great asset, but it makes exploring the design space computationally expensive. Herein, we present how a constrained Bayesian optimization algorithm can be used to efficiently design a criticality experiment. It uses Gaussian processes as a surrogate model to probe the design space and to reduce the number of code executions that are needed to find the optimum. We demonstrate constrained Bayesian optimization with a Pu-239/polyethylene solution system and a TEX experiment that is designed for critical ity safety validation of a nuclear waste model at the Hanford Site. For both systems, a global optimum was found within 75 Monte Carlo simulations.

42 ENGINEERING↗

The Kalai-Smorodinsky solution for many-objective Bayesian optimization

An ongoing aim of research in multiobjective Bayesian optimization is to extend its applicability to a large number of objectives. While coping with a limited budget of evaluations, recovering the set of optimal compromise solutions generally requires numerous observations and is less interpretable since this set tends to grow larger with the number of objectives. We thus propose to focus on a specific solution originating from game theory, the Kalai-Smorodinsky solution, which possesses attractive properties. In particular, it ensures equal marginal gains over all objectives. We further make it insensitive to a monotonic transformation of the objectives by considering the objectives in the copula space. A novel tailored algorithm is proposed to search for the solution, in the form of a Bayesian optimization algorithm: sequential sampling decisions are made based on acquisition functions that derive from an instrumental Gaussian process prior. Our approach is tested on four problems with respectively four, six, eight, and nine objectives. The method is available in the R package GPGame available on CRAN at https://cran.r-project.org/package=GPGame.

Game theory↗

Constrained Bayesian Optimization of Criticality Experiments [Slides]

The design of criticality experiments is typically an iterative process that employs a Monte Carlo transport code. The goal is to find a design that optimizes some variable, like the sensitivity of a response to a cross section, while simultaneously ensuring criticality. The high fidelity of the Monte Carlo code is a great asset, but it makes exploring the design space computationally expensive. Herein, we present how a constrained Bayesian optimization algorithm can be used to efficiently design a criticality experiment. It uses Gaussian processes as a surrogate model to probe the design space and to reduce the number of code executions that are needed to find the optimum. We demonstrate constrained Bayesian optimization with a Pu-239/polyethylene solution system and a TEX experiment that is designed for criticality safety validation of a nuclear waste model at the Hanford Site. For both systems, a global optimum was found within 75 Monte Carlo simulations.

42 ENGINEERING↗

Constrained Bayesian optimization of criticality experiments

The design of criticality experiments is typically an iterative process that employs a Monte Carlo transport code. The goal is to find a design that optimizes some variable, like the sensitivity of a response to a cross section, while simultaneously ensuring criticality. The high fidelity of the Monte Carlo code is a great asset, but it makes exploring the design space computationally expensive. Herein, we present how a constrained Bayesian optimization algorithm can be used to efficiently design a criticality experiment. It uses Gaussian processes as a surrogate model to probe the design space and to reduce the number of code executions that are needed to find the optimum. Furthermore, we demonstrate constrained Bayesian optimization with a Pu-239/polyethylene solution system and a TEX experiment that is designed for criticality safety validation of a nuclear waste model at the Hanford Site. For both systems, a global optimum was found within 75 Monte Carlo simulations.

42 ENGINEERING↗

Modeling Stochastic Variability in Multiband Time-series Data

In preparation for the era of time-domain astronomy with upcoming large-scale surveys, we propose a state-space representation of a multivariate damped random walk process as a tool to analyze irregularly-spaced multifilter light curves with heteroscedastic measurement errors. We adopt a computationally efficient and scalable Kalman filtering approach to evaluate the likelihood function, leading to maximum O(k 3 n) complexity, where k is the number of available bands and n is the number of unique observation times across the k bands. This is a significant computational advantage over a commonly used univariate Gaussian process that can stack up all multiband light curves in one vector with maximum O(k 3 n 3 ) complexity. Using such efficient likelihood computation, we provide both maximum likelihood estimates and Bayesian posterior samples of the model parameters. Three numerical illustrations are presented: (i) analyzing simulated five-band light curves for a comparison with independent single-band fits; (ii) analyzing five-band light curves of a quasar obtained from the Sloan Digital Sky Survey Stripe 82 to estimate short-term variability and timescale; (iii) analyzing gravitationally lensed g- and r-band light curves of Q0957+561 to infer the time delay. Two R packages, Rdrw and timedelay, are publicly available to fit the proposed models.

79 ASTRONOMY AND ASTROPHYSICS↗

Effective Opacity of the Intergalactic Medium from Galaxy Spectra Analysis

We measure the effective opacity (τ{sub eff}) of the intergalactic medium from the composite spectra of 281 Lyman-break galaxies in the redshift range 2 ≲ z ≲ 3. Our spectra are taken from the COSMOS Lyα Mapping And Tomographic Observations survey derived from the Low Resolution Imaging Spectrometer on the W.M. Keck I telescope. We generate composite spectra in two redshift intervals and fit them with spectral energy distribution (SED) models composed of simple stellar populations. Extrapolating these SED models into the Lyα forest, we measure the effective Lyα opacity (τ{sub eff}) in the 2.02 ≤ z ≤ 2.44 range. At z = 2.22, we estimate τ{sub eff} =0.159±0.001 from a power-law fit to the data. These measurements are consistent with estimates from quasar analyses at z < 2.5 indicating that the systematic errors associated with normalizing quasar continua are not substantial. We provide a Gaussian processes model of our results and previous τ{sub eff} measurements that describes the steep redshift evolution in τ{sub eff} from z = 1.5–4.

79 ASTRONOMY AND ASTROPHYSICS↗

General Multifidelity Surrogate Models: Framework and Active-Learning Strategies for Efficient Rare Event Simulation

Estimating the probability of failure for complex real-world systems using high-fidelity computational models is often prohibitively expensive, especially when the probability is small. Exploiting low-fidelity models can make this process more feasible, but merging information from multiple low-fidelity and high-fidelity models poses several challenges. Here, this paper presents a robust multi-fidelity surrogate modeling strategy in which the multi-fidelity surrogate is assembled using an active learning strategy using an on-the-fly model adequacy assessment set within a subset simulation framework for efficient reliability analysis. The multi-fidelity surrogate is assembled by first applying a Gaussian process correction to each low-fidelity model and assigning a model probability based on the model's local predictive accuracy and cost. Three strategies are proposed to fuse these individual surrogates into an overall surrogate model based on model averaging and deterministic/stochastic model selection. The strategies also dictate which model evaluations are necessary. No assumptions are made about the relationships between low-fidelity models, while the high-fidelity model is assumed to be the most accurate and most computationally expensive model. Through two analytical and two numerical case studies, including a case study evaluating the failure probability of Tristructural isotropic-coated (TRISO) nuclear fuels, the algorithm is shown to be highly accurate while drastically reducing the number of high-fidelity model calls (and hence computational cost).

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Emulator-based Bayesian calibration of a subglacial drainage model

Subglacial drainage models, often motivated by the relationship between hydrology and ice flow, sensitively depend on numerous unconstrained parameters. We explore using borehole water-pressure time series to calibrate the uncertain parameters of a popular subglacial drainage model, taking a Bayesian perspective to quantify the uncertainty in parameter estimates and in the calibrated model predictions. To reduce the computation time associated with Markov Chain Monte Carlo sampling, we construct a fast Gaussian process emulator to stand in for the subglacial drainage model. We first carry out a calibration experiment using synthetic observations consisting of model simulations with hidden parameter values as a demonstration of the method. Using real borehole water pressures measured in western Greenland, we find meaningful constraints on four of the eight model parameters and a factor-of-three reduction in uncertainty of the calibrated model predictions. These experiments illustrate Gaussian process-based Bayesian inference as a useful tool for calibration and uncertainty quantification of complex glaciological models using field data. However, significant differences between the calibrated model and the borehole data suggest that structural limitations of the model, rather than poorly constrained parameters or computational cost, remain the most important constraint on subglacial drainage modelling.

58 GEOSCIENCES↗

ELM: AN ALGORITHM TO ESTIMATE THE ALPHA ABUNDANCE FROM LOW-RESOLUTION SPECTRA

We have investigated a novel methodology using the extreme learning machine (ELM) algorithm to determine the α abundance of stars. Applying two methods based on the ELM algorithm—ELM+spectra and ELM+Lick indices—to the stellar spectra from the ELODIE database, we measured the α abundance with a precision better than 0.065 dex. By applying these two methods to the spectra with different signal-to-noise ratios (S/Ns) and different resolutions, we found that ELM+spectra is more robust against degraded resolution and ELM+Lick indices is more robust against variation in S/N. To further validate the performance of ELM, we applied ELM+spectra and ELM+Lick indices to SDSS spectra and estimated α abundances with a precision around 0.10 dex, which is comparable to the results given by the SEGUE Stellar Parameter Pipeline. We further applied ELM to the spectra of stars in Galactic globular clusters (M15, M13, M71) and open clusters (NGC 2420, M67, NGC 6791), and results show good agreement with previous studies (within 1σ). A comparison of the ELM with other widely used methods including support vector machine, Gaussian process regression, artificial neural networks, and linear least-squares regression shows that ELM is efficient with computational resources and more accurate than other methods.

79 ASTRONOMY AND ASTROPHYSICS↗

Bayesian inference for high confidence signal validation and sensor calibration assessment - 328

Online Monitoring (OLM) systems has been investigated for applications in the nuclear industry for the past two decades. These systems offer economic and technological advantages over the currently implemented time based testing methods for sensors and associated instrumentations. The advantages offered not only augment economic viability of current nuclear reactors but will be an inevitable change called for to suit the technologies of the advanced reactors (Gen III+). Irrespective, the acceptance of OLM into the regulatory structure in the U.S. is still hindered by technical and structural clarity on uncertainty quantification used by these systems. A Bayesian inference technique is implemented in the form of a residual model to quantify uncertainty in OLM predictions. The developed system used Gaussian processes to represent the OLM predictions and model inadequacy, the stationary representation of model based uncertainty. In addition, model selection techniques are used in conjunction with numerical methods such as Genetic Algorithm (GA) and Gibb's sampler to add robustness to the estimation model inadequacy, particularly in the presence of drifts and faulted conditions. The prediction model developed herein find applicability in fault detection, obtaining real values of processes after the onset of faults etc. Further development looks into the extension of current method for a multiple output scale implementation. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Bayesian Framework for Multi-Timescale State Estimation in Low-Observable Distribution Systems

To support the smart grid paradigm, there has been a significant increase in sensor deployments and metering infrastructure in distribution systems. However, the measurements provided by these sensors and metering devices are typically sampled at different rates and could suffer from losses during the aggregation process. It is crucial to effectively reconcile the time-series measurements for a reliable state estimation. While weighted least squares has been the traditional approach for state estimation, sparsity-based approaches like matrix completion have become popular due to their superior performance in low-observability conditions. This paper proposes a Bayesian framework for both multi-timescale data aggregation and matrix completion based state estimation. Specifically, the multiscale time-series data aggregated from heterogenous sources are reconciled using a multitask Gaussian process that exploits the spatio-temporal correlations. Here, the resulting consistent timeseries alongwith the confidence bound on the imputations are fed into a Bayesian matrix completion method augmented with linearized power-flow constraints to accurately estimate the states in low-observability conditions. Results on three phase unbalanced IEEE 37 and IEEE 123 bus test systems reveal the superior performance of the proposed Bayesian framework. The computational complexity for the proposed Bayesian framework is also quantified.

42 ENGINEERING↗

Solving Stochastic Inverse Problems for Property–Structure Linkages Using Data-Consistent Inversion and Machine Learning

Determining process–structure–property linkages is one of the key objectives in material science, and uncertainty quantification plays a critical role in understanding both process–structure and structure–property linkages. In this work, we seek to learn a distribution of microstructure parameters that are consistent in the sense that the forward propagation of this distribution through a crystal plasticity finite element model matches a target distribution on materials properties. This stochastic inversion formulation infers a distribution of acceptable/consistent microstructures, as opposed to a deterministic solution, which expands the range of feasible designs in a probabilistic manner. Furthermore, to solve this stochastic inverse problem, we employ a recently developed uncertainty quantification framework based on push-forward probability measures, which combines techniques from measure theory and Bayes’ rule to define a unique and numerically stable solution. This approach requires making an initial prediction using an initial guess for the distribution on model inputs and solving a stochastic forward problem. To reduce the computational burden in solving both stochastic forward and stochastic inverse problems, we combine this approach with a machine learning Bayesian regression model based on Gaussian processes and demonstrate the proposed methodology on two representative case studies in structure–property linkages.

36 MATERIALS SCIENCE↗

Sensitivity study of a full-scale industrial spray-injected fluidized bed reactor

The industrial fluidized bed reactor (FBR) described here is designed to convert an aqueous solid laden stream into a consistent solid product material. Additionally, the FBR is heated to 650 °C and the aqueous reactant is spray-atomized into the fluidized bed, converting it into a granular solid through a series of heterogeneous chemical reactions. The objective of this study is to use the MFiX two-fluid model as a screening tool to investigate the effects of off-nominal conditions for the purpose of establishing performance guidelines. This information will be helpful to assess whether the identified boundaries of the system are steep or gradual. Initially, 13 independent operating parameters were sampled in a one-at-a-time manner using high, low, and base values and ranked against a set of performance criteria. Five operating parameters were found to have the largest effect (bed particle size, bed particle density, coal particle size, spray feed flow rate, and fluidizing gas flow rate) on three quantities of interest—bed differential temperature, solids velocity, and bed voidage. Latin hypercube sampling was used to generate a minimal number of random values for various combinations of input parameters. The spray feed flow rate was the most significant parameter for the temperature differences, and the coal particle and bed particle size were significant contributors. The bed particle size had the largest effect on the low velocity of the bed particles, with the coal particle size as a contributing second effect. The high solid packing shows the coal particle size with the largest effect, and significant secondary contributions from the feed flow rate and fluidizing gas flow rate. Finally, the fits to the five-dimensional Gaussian Process models were 0.7797, 0.8664, and 0.9440 for the temperature, velocity, and solids packing respectively.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Combining Observational and Computational Uncertainty in Calibration Experiments

Traditional computer experiments using emulators have treat the emulator output as containing no error; however, this is not always feasible, as there exist computational codes which produce stochastic output, and future computations may require a balance between the time or resources to run the code and the accuracy of the output. This paper proposes a framework for incorporating both computational uncertainty and observational uncertainty into an analysis using a Gaussian process regression and calibration, and provides an example using simulated output from the one-dimensional diffusion equation. The particular values of observational and computation uncertainty which minimize a cost function are obtained.

97 MATHEMATICS AND COMPUTING↗

Radial-velocity Precision of ESPRESSO Through the Analysis of the Solar Twin HIP 11915

Different stellar phenomena affect radial velocities (RVs), causing variations large enough to make it difficult to identify planet signals from the stellar variability. RV variations caused by stellar oscillations and granulation can be reduced through some methods, but the impact of rotationally modulated magnetic activity on RV, due to stellar active regions, is harder to correct. New instrumentation promises an improvement in precision of one order of magnitude, from about 1 m s{sup -1} to about 10 cm s{sup -1}. In this context, we report our first results from 24 spectroscopic Echelle SPectrograph for Rocky Exoplanets and Stable Spectroscopic Observations (ESPRESSO)/Very-Large Telescope observations of the solar twin star HIP 11915, spread over 60 nights. We used a Gaussian Process approach and found for HIP 11915 a RV residual rms scatter of about 20 cm s{sup -1}, representing an upper limit for the performance of ESPRESSO.

47 OTHER INSTRUMENTATION↗