Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Bayesian Robustness”

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 253 records · Page 14

A Full 24-Parameter MSSM Exploration

Up until now a complete scan in all phenomenologically relevant directions of the MSSM at the TeV scale for performing global fit has not been done. Given the imminent start of operation of the LHC, this is a major gap on our quest to discovering and understanding the physical implications of low energy supersymmetry. The main reason for this is the large number of parameters involved that makes it computationally extremely expensive using the traditional methods. In this talk I demonstrate that with advanced Bayesian sampling techniques the problem is solvable. The results from the explored 24-parameter TeV scale MSSM (phenoMSSM) are remarkably distinct from previous studies and are independent of models for supersymmetry breaking and mediation mechanisms. Hence they are a more robust guide to searches for supersymmetry.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

A generalized muon trajectory estimation algorithm with energy loss for application to muon tomography

Here, this work presents a generalized muon trajectory estimation (GMTE) algorithm to estimate the path of a muon in either uniform or nonuniform media. The use of cosmic ray muons in nuclear nonproliferation and safeguards verification applications has recently gained attention due to the non-intrusive and passive nature of the inspection, penetrating capabilities, as well as recent advances in detectors that measure position and direction of the individual muons before and after traversing the imaged object. However, muon image reconstruction techniques are limited in resolution due to low muon flux and the effects of multiple Coulomb scattering (MCS). Current reconstruction algorithms, e.g., point of closest approach (PoCA) or straight-line path (SLP), rely on overly simple assumptions for muon path estimation through the imaged object. For robust muon tomography, efficient and flexible physics-based algorithms are needed to model the MCS process and accurately estimate the most probable trajectory of a muon as it traverses an object. In the present work, the use of a Bayesian framework and a Gaussian approximation of MCS are explored for estimation of the most likely path of a cosmic ray muon traversing uniform or nonuniform media and undergoing MCS. The algorithm’s precision is compared to Monte Carlo simulated muon trajectories. It was found that the algorithm is expected to be able to predict muon tracks to less than 1.5 mm RMS for 0.5 GeV muons and 0.25 mm RMS for 3 GeV muons, a 50% improvement compared to SLP and 15% improvement when compared to PoCA. Further, a 30% increase in useful muon flux was observed relative to PoCA. Muon track prediction improved for higher muon energies or smaller penetration depth where energy loss is not significant. Finally, the effect of energy loss due to ionization is investigated, and a linear energy loss relation that is easy to use is proposed.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Power generation forecasting for solar plants based on Dynamic Bayesian networks by fusing multi-source information

A Dynamic Bayesian network (DBN) model for solar power generation forecasting in solar plants is proposed in this paper. The key idea is to fuse sensor data, operational indicators, meteorological data, lagged output power information, and model errors for more accurate short-term (e.g., hours) and mid-term (e.g., days to weeks) power generation forecasting. The proposed DBN augments automated data-driven structure learning with expert knowledge encoding using continuous and categorical data given constraints to represent causal relationships within a solar inverter system. Additionally, an error compensation mechanism is proposed to capture temporal fluctuation. The effectiveness of the DBN on solar power generation forecasting was evaluated by rolling window analysis with one-year testing data collected from a local solar plant. The proposed DBN is compared with four state-of-art methods including support-vector regression (SVR), k-nearest neighbors (kNN), artificial neural network (ANN), and long short-term memory (LSTM) models. The result show that the proposed DBN achieves better accuracy in general, and it is not as data-hungry as some neural network-based models. The proposed DBN is also shown to have robust and consistent forecasting power with different forecasting horizons. The accuracy is 92% - 95% from one hour to one week ahead forecasting.

14 SOLAR ENERGY↗

Methods for Incorporating Model Uncertainty into Exoplanet Atmospheric Analysis

A key goal of exoplanet spectroscopy is to measure atmospheric properties, such as abundances of chemical species, in order to connect them to our understanding of atmospheric physics and planet formation. In this new era of high-quality JWST data, it is paramount that these measurement methods are robust. When comparing atmospheric models to observations, multiple candidate models may produce reasonable fits to the data. Typically, conclusions are reached by selecting the best-performing model according to some metric. This ignores model uncertainty in favor of specific model assumptions, potentially leading to measured atmospheric properties that are overconfident and/or incorrect. In this paper, we compare three ensemble methods for addressing model uncertainty by combining posterior distributions from multiple analyses: Bayesian model averaging, a variant of Bayesian model averaging using leave-one-out predictive densities, and stacking of predictive distributions. We demonstrate these methods by fitting the Hubble Space Telescope (HST) + Spitzer transmission spectrum of the hot Jupiter HD 209458b using models with different cloud and haze prescriptions. All of our ensemble methods lead to uncertainties on retrieved parameters that are larger but more realistic and consistent with physical and chemical expectations. Since they have not typically accounted for model uncertainty, uncertainties of retrieved parameters from HST spectra have likely been underreported. We recommend stacking as the most robust model combination method. Our methods can be used to combine results from independent retrieval codes and from different models within one code. They are also widely applicable to other exoplanet analysis processes, such as combining results from different data reductions.

79 ASTRONOMY AND ASTROPHYSICS↗

Inflation with an extra light scalar field after Planck

Bayesian inference techniques are used to investigate situations where an additional light scalar field is present during inflation and reheating. This includes (but is not limited to) curvaton-type models. We design a numerical pipeline where ≅ 200 inflaton setups × 10 reheating scenarios = 2000 models are implemented and we present the results for a few prototypical potentials. We find that single-field models are remarkably robust under the introduction of light scalar degrees of freedom. Models that are ruled out at the single-field level are not improved in general, because good values of the spectral index and the tensor-to-scalar ratio can only be obtained for very fine-tuned values of the extra field parameters and/or when large non-Gaussianities are produced. The only exception is quartic large-field inflation, so that the best models after Planck are of two kinds: plateau potentials, regardless of whether an extra field is added or not, and quartic large-field inflation with an extra light scalar field, in some specific reheating scenarios. Using Bayesian complexity, we also find that more parameters are constrained for the models we study than for their single-field versions. This is because the added parameters not only contribute to the reheating kinematics but also to the cosmological perturbations themselves, to which the added field contributes. The interplay between these two effects lead to a suppression of degeneracies that is responsible for having more constrained parameters.

79 ASTRONOMY AND ASTROPHYSICS↗

In Situ Inference for Earth System Predictability

An understanding of future evolution in precipitation extremes is critical to numerous DOE mission questions. Extreme events are by nature short time-scale events that are difficult to diagnose in available model data. Accurate modeling of extreme events necessarily requires high spatial resolution at the storm scale locally. However, the environment in which storms grow is dependent on global, remote, processes. These complex spatiotemporal relationships are impossible to diagnose at resolutions required to accurately model storms responsible for extreme precipitation. At exascale, climate simulations will produce results at fine enough resolution to investigate these relationships. However, the resulting data from these simulations will be far too large to save for post-simulation analysis. We advocate for fitting statistical models inside the simulations as they run, a context known as in situ, which will facilitate scientific investigations using the full fine-scale data stream. Figure 1 shows an example of the type of model we could consider, a Bayesian hierarchical spatial regression model. Precipitation extremes at each grid cell are modeled using extreme value distributions. Since extremes are rare, fitting models to individual grid cells can result in high variance and poor estimates. Instead, the model can be made more robust by smoothing the parameters of the extreme value model across space. Additionally, the parameters themselves can be functionally linked to other variables elsewhere in the simulation. Thus, we can use the fine-scale data to build more robust models for extremes that link extreme behavior to other climate patterns.

54 ENVIRONMENTAL SCIENCES↗

In Situ Inference for Earth System Predictability

Focal Area: Focal Area 3: Insight gleaned from complex simulated data using AI, big data analytics, and other advanced methods, including explainable AI and physics- or knowledge-guided AI. Science Challenge: An understanding of future evolution in precipitation extremes is critical to numerous DOE mission questions. Extreme events are by nature short time-scale events that are difficult to diagnose in available model data. Accurate modeling of extreme events necessarily requires high spatial resolution at the storm scale locally. However, the environment in which storms grow is dependent on global, remote, processes. These complex spatiotemporal relationships are impossible to diagnose at resolutions required to accurately model storms responsible for extreme precipitation. At exascale, climate simulations will produce results at fine enough resolution to investigate these relationships. However, the resulting data from these simulations will be far too large to save for post-simulation analysis. We advocate for fitting statistical models inside the simulations as they run, a context known as in situ, which will facilitate scientific investigations using the full fine-scale data stream. Figure 1 shows an example of the type of model we could consider, a Bayesian hierarchical spatial regression model. Precipitation extremes at each grid cell are modeled using extreme value distributions. Since extremes are rare, fitting models to individual grid cells can result in high variance and poor estimates. Instead, the model can be made more robust by smoothing the parameters of the extreme value model across space. Additionally, the parameters themselves can be functionally linked to other variables elsewhere in the simulation. Thus, we can use the fine-scale data to build more robust models for extremes that link extreme behavior to other climate patterns.

54 ENVIRONMENTAL SCIENCES↗

Sequential ensemble transform for Bayesian inverse problems

In this work, we present the Sequential Ensemble Transform (SET) method, an approach for generating approximate samples from a Bayesian posterior distribution. The method explores the posterior distribution by solving a sequence of discrete optimal transport problems to produce a series of transport plans which map prior samples to posterior samples. We prove that the sequence of Dirac mixture distributions produced by the SET method converges weakly to the true posterior as the sample size approaches infinity. Furthermore, our numerical results indicate that, when compared to standard Sequential Monte Carlo (SMC) methods, the SET approach is more robust to the choice of Markov mutation kernels and requires less computational efforts to reach a similar accuracy when used to explore complex posterior distributions. Finally, we describe adaptive schemes that allow to completely automate the use of the SET method.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Bayesian probabilistic model for life prediction and fault mode classification of solid state luminaires

A new method has been developed for assessment of the onset of degradation in solid state luminaires to classify failure mechanisms by using metrics beyond lumen degradation that are currently used for identification of failure. Luminous Flux output, Correlated Color Temperature Data on Philips LED Lamps has been gathered under 85°C/85%RH till lamp failure. Failure modes of the test population of the lamps have been studied to understand the failure mechanisms in 85°C/85%RH accelerated test. Results indicate that the dominant failure mechanism is the discoloration of the LED encapsulant inside the lamps which is the likely cause for the luminous flux degradation and the color shift. The acquired data has been used in conjunction with Bayesian Probabilistic Models to identify luminaires with onset of degradation much prior to failure through identification of decision boundaries between lamps with accrued damage and lamps beyond the failure threshold in the feature space. In addition luminaires with different failure modes have been classified separately from healthy pristine luminaires. The α-λ plots have been used to evaluate the robustness of the proposed methodology. Results show that the predicted degradation for the lamps tracks the true degradation observed during 85°C/85%RH during accelerated life test fairly closely within the ±20% confidence bounds. Correlation of model prediction with experimental results indicates that the presented methodology allows the early identification of the onset of failure much prior to development of complete failure distributions and can be used for assessing the damage state of SSLs in fairly large deployments. It is expected that, the new prediction technique will allow the development of failure distributions without testing till L70 life for the manifestation of failure.

Bayesian probability↗

Consequential improvement acquisition function for efficient multi-fidelity Bayesian optimization

Abstract Surrogate-based Bayesian optimization has been widely applied in design optimization to increase sampling efficiency. However, the cost for each evaluation of the objective function can still be very high when physical experiments or large-scale simulations are involved. Multi-fidelity Bayesian optimization is the new approach to further improve the sampling efficiency by reducing the number of expensive samples at the highest fidelity level and supplementing them with less expensive ones at low-fidelity levels. In this paper, a new consequential improvement (CI) acquisition function is proposed to allow for the simultaneous selection of the solution and the fidelity level in problems with a known hierarchy of fidelity levels. The new CI acquisition function incorporates the consequential effectiveness of objective improvement with the considerations of cost, accuracy, and validity differences between high- and low-fidelity samples in engineering practice. The new method of multi-fidelity Bayesian optimization based on the CI is demonstrated with several analytical and simulation-based design examples. In the simulation-based design optimization example, the results show that the CI acquisition function has a decisive advantage in the sampling efficiency over the other methods of multi-fidelity Bayesian optimization with simultaneous selection. The results indicate that the proposed method is particularly advantageous in solving high-dimensional problems and when large cost ratios between high- and low-fidelity evaluations exist and high-fidelity validation is mandatory. Furthermore, the method robustly avoids the prevalent issue of over sampling at low-fidelity levels.

Aydogdu, Ibrahim [Georgia Institute of Technology,↗

Evaluation of kriging based surrogate models constructed from mesoscale computations of shock interaction with particles

Macro-scale computations of shocked particulate flows require closure laws that model the exchange of momentum/energy between the fluid and particle phases. Closure laws are constructed in this work in the form of surrogate models derived from highly resolved mesoscale computations of shock-particle interactions. The mesoscale computations are performed to calculate the drag force on a cluster of particles for different values of Mach Number and particle volume fraction. Two Kriging-based methods, viz. the Dynamic Kriging Method (DKG) and the Modified Bayesian Kriging Method (MBKG) are evaluated for their ability to construct surrogate models with sparse data; i.e. using the least number of mesoscale simulations. It is shown that if the input data is noise-free, the DKG method converges monotonically; convergence is less robust in the presence of noise. The MBKG method converges monotonically even with noisy input data and is therefore more suitable for surrogate model construction from numerical experiments. This work is the first step towards a full multiscale modeling of interaction of shocked particle laden flows.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Constraining neutron-star matter with microscopic and macroscopic collisions

Interpreting high-energy, astrophysical phenomena, such as supernova explosions or neutron-star collisions, requires a robust understanding of matter at supranuclear densities. However, our knowledge about dense matter explored in the cores of neutron stars remains limited. Fortunately, dense matter is not probed only in astrophysical observations, but also in terrestrial heavy-ion collision experiments. Here we use Bayesian inference to combine data from astrophysical multi-messenger observations of neutron stars and from heavy-ion collisions of gold nuclei at relativistic energies with microscopic nuclear theory calculations to improve our understanding of dense matter. We find that the inclusion of heavy-ion collision data indicates an increase in the pressure in dense matter relative to previous analyses, shifting neutron-star radii towards larger values, consistent with recent observations by the Neutron Star Interior Composition Explorer mission. Our findings show that constraints from heavy-ion collision experiments show a remarkable consistency with multi-messenger observations and provide complementary information on nuclear matter at intermediate densities. This work combines nuclear theory, nuclear experiment and astrophysical observations, and shows how joint analyses can shed light on the properties of neutron-rich supranuclear matter over the density range probed in neutron stars.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

ASTROMETRY.NET: BLIND ASTROMETRIC CALIBRATION OF ARBITRARY ASTRONOMICAL IMAGES

We have built a reliable and robust system that takes as input an astronomical image, and returns as output the pointing, scale, and orientation of that image (the astrometric calibration or World Coordinate System information). The system requires no first guess, and works with the information in the image pixels alone; that is, the problem is a generalization of the 'lost in space' problem in which nothing-not even the image scale-is known. After robust source detection is performed in the input image, asterisms (sets of four or five stars) are geometrically hashed and compared to pre-indexed hashes to generate hypotheses about the astrometric calibration. A hypothesis is only accepted as true if it passes a Bayesian decision theory test against a null hypothesis. With indices built from the USNO-B catalog and designed for uniformity of coverage and redundancy, the success rate is >99.9% for contemporary near-ultraviolet and visual imaging survey data, with no false positives. The failure rate is consistent with the incompleteness of the USNO-B catalog; augmentation with indices built from the Two Micron All Sky Survey catalog brings the completeness to 100% with no false positives. We are using this system to generate consistent and standards-compliant meta-data for digital and digitized imaging from plate repositories, automated observatories, individual scientific investigators, and hobbyists. This is the first step in a program of making it possible to trust calibration meta-data for astronomical data of arbitrary provenance.

07 ISOTOPE AND RADIATION SOURCES↗

Constraints on the initial-final mass relation from wide double white dwarfs

We present observational constraints on the initial-final mass relation (IFMR) using wide double white dwarfs (DWDs). We identify 65 new candidate wide DWDs within the Sloan Digital Sky Survey, bringing the number of candidate wide DWDs to 142. We then engage in a spectroscopic follow-up campaign and collect existing spectra for these objects; using these spectra, we derive masses and cooling ages for 54 hydrogen (DA) WDs in DWDs. We also identify one new DA/DB pair, four candidate DA/DC pairs, four candidate DA/DAH pairs, and one new candidate triple degenerate system. Because wide DWDs are co-eval and evolve independently, the difference in the pre-WD lifetimes should equal the difference in the WD cooling ages. We use this to develop a Bayesian hierarchical framework and construct a likelihood function to determine the probability that any particular IFMR fits a sample of wide DWDs. We then define a parametric model for the IFMR and find the best parameters indicated by our sample of DWDs. We place robust constraints on the IFMR for initial masses of 2–4 M{sub ⊙}. The WD masses produced by our model for stars within this mass range differ from those predicted by semi-empirical fits to open cluster WDs. Within this mass range, where there are few constraining open cluster WDs and disagreements in the cluster ages, wide DWDs may provide more reliable constraints on the IFMR. Expanding this method to the many wide DWDs expected to be discovered by Gaia may transform our understanding of the IFMR.

79 ASTRONOMY AND ASTROPHYSICS↗

Characterization and automated optimization of laser-driven proton beams from converging liquid sheet jet targets

Compact, stable, and versatile laser-driven ion sources hold great promise for applications ranging from medicine to materials science and fundamental physics. While single-shot sources have demonstrated favorable beam properties, including the peak fluxes necessary for several applications, high-repetition-rate operation will be necessary to generate and sustain the high average flux needed for many of the most exciting applications of laser-driven ion sources. Further, to navigate through the high-dimensional space of laser and target parameters toward experimental optima, it is essential to develop ion acceleration platforms compatible with machine learning techniques and capable of autonomous real-time optimization. Here, we present a multi-Hz ion acceleration platform employing a liquid sheet jet target. We characterize the laser-plasma interaction and the laser-driven proton beam across a variety of key parameters governing the interaction using an extensive suite of online diagnostics. We also demonstrate real-time, closed-loop optimization of the ion beam maximum energy by tuning the laser wave front using a Bayesian optimization scheme. This approach increased the maximum proton energy by 11% compared to a manually optimized wave front by enhancing the energy concentration within the laser focal spot, demonstrating the potential for closed-loop optimization schemes to tune future ion accelerators for robust high-repetition-rate operation.

Glenn, G. D. [SLAC National Accelerator Laboratory↗

Optimal sizing of battery energy storage systems for peak shaving and demand response using a degradation-aware Bayesian Optimization-Mixed-Integer Linear Programming framework

The increasing integration of renewable energy and rising electricity demand highlight the importance of battery energy storage systems for peak shaving and demand response. Unlike prior approaches that overlook operational impacts on degradation, this study proposes a Bayesian Optimization–Mixed Integer Linear Programming framework for optimal battery energy storage system sizing. In this framework, Mixed Integer Linear Programming determines short-term scheduling while a calibrated electrochemical model iteratively evaluates degradation. The central hypothesis is that the framework can efficiently identify optimal sizes that yield realistic and economically robust outcomes. The method is tested across three scenarios: peak shaving, peak shaving with energy-reduction demand response, and peak shaving with power-reduction demand response. Results show that the framework converge to the optimum within 20 iterations out of 150 possible sizes. Under baseline conditions, the framework consistently selects the smallest feasible system, minimizing unnecessary degradation costs from oversized storage. Sensitivity analyses reveal that larger systems are favored as demand rates or incentives increase. Comparisons of demand response programs indicate that power-reduction demand response offers greater economic benefits than energy-reduction demand response, although demand savings from peak shaving remain the dominant contributor to overall performance. This study demonstrates that the proposed framework balances computational tractability with degradation fidelity, identifies critical economic thresholds for investment, and offers a practical, flexible tool to guide industrial stakeholders in cost-effective battery energy storage system deployment.

Batteries↗

Uranium particle age dating, aggregation, and model age best estimators

We present important aspects of uranium particle age dating by Large-Geometry Secondary Ion Mass Spectrometry (LG-SIMS) that can introduce bias and increase model age uncertainties, especially for small, young, and/or low-enriched particles. This metrology is important for applications related to International Nuclear Safeguards. We explore influential factors related to model age estimation, including the effects of evolving surface chemistry on inter-element measurements of particles (e.g., Th and U), detector background, and aggregation methods using simulated and actual particle samples. We introduce a new model age estimator, called “mid68”, that supplements 95% confidence intervals, providing a “best estimate” and uncertainty about the most likely age. The mid68 estimator can be calculated using the Feldman and Cousins method or Bayesian methods and provides a value with a symmetric uncertainty that can be used for calculations and approximate aggregation of processed model age values when the raw data and correction factors are not available. For particles yielding low 230 Th counts amidst nonzero detector background, their underlying model age probability distributions are asymmetric, so the mid68 estimator provides additional robust information regarding the underlying model age likelihood. This study provides a comprehensive and timely examination of critical aspects of uranium particle age dating as more laboratories establish particle chronometry capabilities.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Redshift inference from the combination of galaxy colours and clustering in a hierarchical Bayesian model

Powerful current and future cosmological constraints using high-precision measurements of the large-scale structure of galaxies and its weak gravitational lensing effects rely on accurate characterization of the redshift distributions of the galaxy samples using only broad-band imaging. In this paper, we present a framework for constraining both the redshift probability distributions of galaxy populations and the redshifts of their individual members. We use a hierarchical Bayesian model (HBM) which provides full posterior distributions on those redshift probability distributions, and, for the first time, we show how to combine survey photometry of single galaxies and the information contained in the galaxy clustering against a well-characterized tracer population in a robust way. One critical approximation turns the HBM into a system amenable to efficient Gibbs sampling. We show that in the absence of photometric information, this method reduces to commonly used clustering redshift estimators. Using a simple model system, we show how the incorporation of clustering information with photo-$z$’s tightens redshift posteriors, and can overcome biases or gaps in the coverage of a spectroscopic prior. The method enables the full propagation of redshift uncertainties into cosmological analyses, and uses all the information at hand to reduce those uncertainties and associated potential biases.

79 ASTRONOMY AND ASTROPHYSICS↗