Results for “model calibration tasks” — Page 9

DOE OSTI2024

Self-Supervised Cloud Classification

Abstract Low-level marine clouds play a pivotal role in Earth’s weather and climate through their interactions with radiation, heat and moisture transport, and the hydrological cycle. These interactions depend on a range of dynamical and microphysical processes that result in a broad diversity of cloud types and spatial structures, and a comprehensive understanding of cloud morphology is critical for continued improvement of our atmospheric modeling and prediction capabilities moving forward. Deep learning has recently accelerated our ability to study clouds using satellite remote sensing, and machine learning classifiers have enabled detailed studies of cloud morphology. A major limitation of deep learning approaches to this problem, however, is the large number of hand-labeled samples that are required for training. This work applies a recently developed self-supervised learning scheme to train a deep convolutional neural network (CNN) to map marine cloud imagery to vector embeddings that capture information about mesoscale cloud morphology and can be used for satellite image classification. The model is evaluated against existing cloud classification datasets and several use cases are demonstrated, including training cloud classifiers with very few labeled samples, interrogation of the CNN’s learned internal feature representations, cross-instrument application, and resilience against sensor calibration drift and changing scene brightness. The self-supervised approach learns meaningful internal representations of cloud structures and achieves comparable classification accuracy to supervised deep learning methods without the expense of creating large hand-annotated training datasets. Significance Statement Marine clouds heavily influence Earth’s weather and climate, and improved understanding of marine clouds is required to improve our atmospheric modeling capabilities and physical understanding of the atmosphere. Recently, deep learning has emerged as a powerful research tool that can be used to identify and study specific marine cloud types in the vast number of images collected by Earth-observing satellites. While powerful, these approaches require hand-labeling of training data, which is prohibitively time intensive. This study evaluates a recently developed self-supervised deep learning method that does not require human-labeled training data for processing images of clouds. We show that the trained algorithm performs competitively with algorithms trained on hand-labeled data for image classification tasks. We also discuss potential downstream uses and demonstrate some exciting features of the approach including application to multiple satellite instruments, resilience against changing image brightness, and its learned internal representations of cloud types. The self-supervised technique removes one of the major hurdles for applying deep learning to very large atmospheric datasets.

54 ENVIRONMENTAL SCIENCES↗

NASA NTRS2016

A Computationally-Efficient, Multi-Mechanism Based Framework for the Comprehensive Modeling of the Evolutionary Behavior of Shape Memory Alloys

The report summarizes the accomplishments made during the 4-year duration of the project. Here, the major emphasis is placed on the different tasks performed by the two research teams; i.e., the modeling activities by the University of Akron (UA) team and the experimental and neutron diffraction studies conducted by the University of Central Florida (UCF) team, during this 4-year period. Further technical details are given in the upcoming sections by UA and UCF for each of the milestones/years (together with the corresponding figures and captions).The project majorly involved the development, validation, and application of a general theoretical model that is capable of capturing the nonlinear hysteretic responses, including pseudoelasticity, shape memory effect, rate-dependency, multi-axiality, asymmetry in tension versus compression response of shape memory alloys. Among the targeted goals for the SMA model was its ability to account for the evolutionary character response (including transient and long term behavior under sustained cycles) for both conventional and high temperature (HT) SMAs, as well as being able to simulate some of the devices which exploit these unique material systems. This required extensive (uniaxial and multi-axial) experiments needed to guide us in calibrating and characterizing the model. Moreover, since the model is formulated on the theoretical notion of internal state variables (ISVs), neutron diffraction experiments were needed to establish the linkage between the micromechanical changes and these ISVs. In addition, the design of the model should allow easy implementation in large scale finite element application to study the behavior of devices making use of these SMA materials under different loading controls. Summary of the activities, progress/achievements made during this period is given below in details for the University of Akron and the University (Section 2.0) of Central Florida (Section 3.0).

Alloys↗

DOE OSTI2025

Huge ensembles – Part 1: Design of ensemble weather forecasts using spherical Fourier neural operators

Abstract. Simulating low-likelihood high-impact extreme weather events in a warming world is a significant and challenging task for current ensemble forecasting systems. While these systems presently use up to 100 members, larger ensembles could enrich the sampling of internal variability. They may capture the long tails associated with climate hazards better than traditional ensemble sizes. Due to computational constraints, it is infeasible to generate huge ensembles (comprised of 1000–10 000 members) with traditional, physics-based numerical models. In this two-part paper, we replace traditional numerical simulations with machine learning (ML) to generate hindcasts of huge ensembles. In Part 1, we construct an ensemble weather forecasting system based on spherical Fourier neural operators (SFNOs), and we discuss important design decisions for constructing such an ensemble. The ensemble represents model uncertainty through perturbed-parameter techniques, and it represents initial condition uncertainty through bred vectors, which sample the fastest-growing modes of the forecast. Using the European Centre for Medium-Range Weather Forecasts Integrated Forecasting System (IFS) as a baseline, we develop an evaluation pipeline composed of mean, spectral, and extreme diagnostics. With large-scale, distributed SFNOs with 1.1 billion learned parameters, we achieve calibrated probabilistic forecasts. As the trajectories of the individual members diverge, the ML ensemble mean spectra degrade with lead time, consistent with physical expectations. However, the individual ensemble members' spectra stay constant with lead time. Therefore, these members simulate realistic weather states during the rollout, and the ML ensemble passes a crucial spectral test in the literature. The IFS and ML ensembles have similar extreme forecast indices, and we show that the ML extreme weather forecasts are reliable and discriminating. These diagnostics ensure that the ensemble can reliably simulate the time evolution of the atmosphere, including low-likelihood high-impact extremes. In Part 2, we generate a huge ensemble initialized each day in summer 2023, and we characterize the simulations of extremes.

Mahesh, Ankur↗

DOE OSTI2024

Evidential Deep Learning: Enhancing Predictive Uncertainty Estimation for Earth System Science Applications

Abstract Robust quantification of predictive uncertainty is a critical addition needed for machine learning applied to weather and climate problems to improve the understanding of what is driving prediction sensitivity. Ensembles of machine learning models provide predictive uncertainty estimates in a conceptually simple way but require multiple models for training and prediction, increasing computational cost and latency. Parametric deep learning can estimate uncertainty with one model by predicting the parameters of a probability distribution but does not account for epistemic uncertainty. Evidential deep learning, a technique that extends parametric deep learning to higher-order distributions, can account for both aleatoric and epistemic uncertainties with one model. This study compares the uncertainty derived from evidential neural networks to that obtained from ensembles. Through applications of the classification of winter precipitation type and regression of surface-layer fluxes, we show evidential deep learning models attaining predictive accuracy rivaling standard methods while robustly quantifying both sources of uncertainty. We evaluate the uncertainty in terms of how well the predictions are calibrated and how well the uncertainty correlates with prediction error. Analyses of uncertainty in the context of the inputs reveal sensitivities to underlying meteorological processes, facilitating interpretation of the models. The conceptual simplicity, interpretability, and computational efficiency of evidential neural networks make them highly extensible, offering a promising approach for reliable and practical uncertainty quantification in Earth system science modeling. To encourage broader adoption of evidential deep learning, we have developed a new Python package, Machine Integration and Learning for Earth Systems (MILES) group Generalized Uncertainty for Earth System Science (GUESS) (MILES-GUESS) ( https://github.com/ai2es/miles-guess ), that enables users to train and evaluate both evidential and ensemble deep learning. Significance Statement This study demonstrates a new technique, evidential deep learning, for robust and computationally efficient uncertainty quantification in modeling the Earth system. The method integrates probabilistic principles into deep neural networks, enabling the estimation of both aleatoric uncertainty from noisy data and epistemic uncertainty from model limitations using a single model. Our analyses reveal how decomposing these uncertainties provides valuable insights into reliability, accuracy, and model shortcomings. We show that the approach can rival standard methods in classification and regression tasks within atmospheric science while offering practical advantages such as computational efficiency. With further advances, evidential networks have the potential to enhance risk assessment and decision-making across meteorology by improving uncertainty quantification, a longstanding challenge. This work establishes a strong foundation and motivation for the broader adoption of evidential learning, where properly quantifying uncertainties is critical yet lacking.

Schreck, John S.↗

DOE OSTI2024

AI-assisted detector design for the EIC (AID(2)E)

Artificial Intelligence is poised to transform the design of complex, large-scale detectors like ePIC at the future Electron Ion Collider. Featuring a central detector with additional detecting systems in the far forward and far backward regions, the ePIC experiment incorporates numerous design parameters and objectives, including performance, physics reach, and cost, constrained by mechanical and geometric limits. This project aims to develop a scalable, distributed AI-assisted detector design for the EIC (AID(2)E), employing state-of-the-art multiobjective optimization to tackle complex designs. Supported by the ePIC software stack and using G EANT 4 simulations, our approach benefits from transparent parameterization and advanced AI features. The workflow leverages the PanDA and iDDS systems, used in major experiments such as ATLAS at CERN LHC, the Rubin Observatory, and sPHENIX at RHIC, to manage the compute intensive demands of ePIC detector simulations. Tailored enhancements to the PanDA system focus on usability, scalability, automation, and monitoring. Ultimately, this project aims to establish a robust design capability, apply a distributed AI-assisted workflow to the ePIC detector, and extend its applications to the design of the second detector (Detector-2) in the EIC, as well as to calibration and alignment tasks. Additionally, we are developing advanced data science tools to efficiently navigate the complex, multidimensional trade-offs identified through this optimization process.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2026

Regularization via f -Divergence: An Application to Multi-Oxide Spectroscopic Analysis

In this paper, we explore the application of convolutional neural networks (CNNs) for predicting the chemical composition of complex geologic samples in a simulated Martian atmospheric environment. Specifically, we aim to characterize oxide weight percentages (wt.%) of rock samples analyzed by remote Laser-Induced Breakdown Spectroscopy (LIBS), framing the problem as a multi-target regression task . Neural networks trained on LIBS spectra are prone to overfitting due to high spectral complexity, limited labeled data, and measurement noise. While regularization is critical for improving generalization, common methods (e.g., ℓ 2 regularization) impose constraints not directly tied to data distribution properties. We propose a novel regularization method based on a specific ƒ-divergence induced by a graph-based estimator, designed to constrain the distributional discrepancy between predictions and targets. This regularizer serves a dual purpose: (a) mitigating overfitting by enforcing a constraint on the distributional difference between predictions and noisy targets, and (b) acting as an auxiliary loss that penalizes large divergences. To enable backpropagation, we develop a differentiable approximation of this particular ƒ-divergence, making the method feasible for neural networks. Experiments on ChemCam and SuperCam LIBS calibration spectra show that mathematical equation-divergence regularization outperforms or matches standard regularization methods (ℓ 1 , ℓ 2 , dropout) and the classical baseline, partial least squares (PLS). Combining ƒ-divergence regularization with standard regularization yields further performance gains, indicating that distributional regularization is useful in this context giving a promising direction for robust model training in planetary science applications. Source code is publicly available at Klein and Li (2025), https://doi.org/10.11578/dc.20250530.7.

58 GEOSCIENCES↗

NASA NTRSDate not supplied

OVERFLOW Analysis of Supersonic Retropropulsion Testing on a Blunt Mars Entry Vehicle Concept

Supersonic retropropulsion (SRP) flow over a Hypersonic Inflatable Aerodynamic Decelerator (HIAD) blunt-body vehicle was simulated using the Overflow Computational Fluid Dynamics (CFD) solver. Two nozzle configurations were tested (1E and 1F) as a subset of the seven total configurations in the Descent System Study (DSS) testing campaign. A generalized, Adaptive Mesh Refinement (AMR) shock capturing and plume refinement technique was developed and calibrated for producing automatic and unique grid systems optimized for any given testing condition. Solution independence from grid resolution was determined by successively increasing grid refinement until asymptotic convergence of the mean aerodynamic loads was observed. Overflow Reynolds-averaged Navier-Stokes (RANS) solutions produced realistic SRP flow phenomena, including Mach disk normal shocks contained in high-thrust plumes and bow shock triple-points. Solutions for the 1E nozzle configuration were steady. Conversely, a subset of the 1F test conditions were unsteady and exhibited periodic, non-sinusoidal expansions and contractions of the streamwise-oriented plume, resulting in unsteady aerodynamic loads on the vehicle. Initial investigations into the effect of turbulence modeling fidelity demonstrated significant differences between the aerodynamic loads simulated with RANS and Detached Eddy Simulation (DES). Multiple flow mechanisms were identified as root causes of these differences, including bow shock shape augmentation and reduced entrainment due to reduced turbulence in the DES simulations. Initial comparisons of pitching moment control authority and aerodynamic drag performance between the tested nozzle configurations demonstrated that the 1E configuration may be more advantageous for the Mars Entry, Descent, and Landing (EDL) task.

Supersonic Retropropulsion↗

NASA NTRSDate not supplied

OVERFLOW Analysis of Supersonic Retropropulsion Testing on a Blunt Mars Entry Vehicle Concept

Supersonic retropropulsion (SRP) flow over a Hypersonic Inflatable Aerodynamic Decelerator (HIAD) blunt-body vehicle was simulated using the Overflow Computational Fluid Dynamics (CFD) solver. Two nozzle configurations were tested (1E and 1F) as a subset of the seven total configurations in the Descent System Study (DSS) testing campaign. A generalized, Adaptive Mesh Refinement (AMR) shock capturing and plume refinement technique was developed and calibrated for producing automatic and unique grid systems optimized for any given testing condition. Solution independence from grid resolution was determined by successively increasing grid refinement until asymptotic convergence of the mean aerodynamic loads was observed. Overflow Reynolds-averaged Navier-Stokes (RANS) solutions produced realistic SRP flow phenomena, including Mach disk normal shocks contained in high-thrust plumes and bow shock triple-points. Solutions for the 1E nozzle configuration were steady. Conversely, a subset of the 1F test conditions were unsteady and exhibited periodic, non-sinusoidal expansions and contractions of the streamwise-oriented plume, resulting in unsteady aerodynamic loads on the vehicle. Initial investigations into the effect of turbulence modeling fidelity demonstrated significant differences between the aerodynamic loads simulated with RANS and Detached Eddy Simulation (DES). Multiple flow mechanisms were identified as root causes of these differences, including bow shock shape augmentation and reduced entrainment due to reduced turbulence in the DES simulations. Initial comparisons of pitching moment control authority and aerodynamic drag performance between the tested nozzle configurations demonstrated that the 1E configuration may be more advantageous for the Mars Entry, Descent, and Landing (EDL) task.

Supersonic Retropropulsion↗

DOE OSTI2026

Comparison of theory with the experimental characterization of the spatial frequency response of interferometers using a binary pseudo-random array sample

Experimental evaluations of the surface height response of an interference microscope using a binary pseudo-random array test sample are compared with a theory based on a Fourier optics model. Measurements of key instrument characteristics, including the illumination, imaging, and obscuring apertures of three different Mirau objectives, support the theoretical calculations. Agreement between experimental and theoretical modeling confirms the predictability of the spatial frequency response for the purpose of specification and optimization of instrument configuration for specific metrology tasks. The results also provide confidence in methods of compensating for the decrease in instrument response with spatial frequency.

Calibration↗

NASA NTRS2002

Acoustic and Laser Doppler Anemometer Results for Confluent, 22-Lobed, and Unique-Lobed Mixer Exhaust Systems for Subsonic Jet Noise Reduction

The objective of this task was to develop a design methodology and noise reduction concepts for high bypass exhaust systems which could be applied to both existing production and new advanced engine designs. Special emphasis was given to engine cycles with bypass ratios in the range of 4:1 to 7:1, where jet mixing noise was a primary noise source at full power takeoff conditions. The goal of this effort was to develop the design methodology for mixed-flow exhaust systems and other novel noise reduction concepts that would yield 3 EPNdB noise reduction relative to 1992 baseline technology. Two multi-lobed mixers, a 22-lobed axisymmetric and a 21-lobed with a unique lobe, were designed. These mixers along with a confluent mixer were tested with several fan nozzles of different lengths with and without acoustic treatment in GEAE's Cell 41 under the current subtask (Subtask C). In addition to the acoustic and LDA tests for the model mixer exhaust systems, a semi-empirical noise prediction method for mixer exhaust system is developed. Effort was also made to implement flowfield data for noise prediction by utilizing MGB code. In general, this study established an aero and acoustic diagnostic database to calibrate and refine current aero and acoustic prediction tools.

Salikuddin, M.↗

NASA NTRSDate not supplied

Toward an IMU-Based Space Suit Motion Capture System

Spacesuits are complex engineering systems that sustain human health and enable performance outside Earth-like environments. These systems must support human mobility and physical workload demands while minimizing injury risk during extravehicular activity (EVA). Future EVA operations on the Lunar surface are expected to be more frequent and require higher physical workloads than previously during the ISS, Shuttle, and Apollo programs. To characterize the workloads and ergonomics needs a suit must support, the kinematics of the space suit must be measured during operationally-relevant tasks in ground analog environments. Kinematics capture of the suit is challenging for traditional optical motion capture (OMC) approaches due to marker occlusion, harsh lighting or environmental conditions, and tests with suit surrogates in outdoor field environments. To this end, engineers at NASA are developing the Augmented Suit Inverse Kinematics (ASIK) system, a complete motion capture method and inverse kinematics solver which relies solely on a network of wireless inertial measurement units (IMUs) attached to the major kinematic segments of the spacesuit. The ASIK modeling language allows for the simple inclusion of probabilistic priors such as suit size and shape or IMU poses. The ASIK system was tested in a 7-subject pilot study. Each subject donned NASA’s new prototype exploration spacesuit in the Active Response Gravity Offload System (ARGOS) facility at Johnson Space Center in Houston, TX. The suits were outfitted with 12 IMUs to estimate lower body and trunk kinematics. The suits were also outfitted with a set of reflective OMC markers, and traditional OMC data was collected and processed. Characterization of the ASIK-derived suit joint angles’ accuracy against an optical motion capture datum will be presented. Discussion of these results, as well as discussion of system calibration and nuances of mathematical observability, will be included.

IMU↗

DOE OSTI2022

Upgrade of Gamma Spectrometry Systems for ORNL TRISO Fuel PIE

Gamma spectrometry is a key element in much of the post-irradiation examination (PIE) work performed under the Advanced Gas Reactor Fuel Development and Qualification (AGR) Program (Demkowicz et al. 2015; Stempien et al. 2021). Gamma spectrometers are integrated into three major capabilities used at the Oak Ridge National Laboratory (ORNL) Irradiated Fuels Examination Laboratory (IFEL) for PIE of tristructural-isotropic (TRISO) coated particles and fuel compacts: the Core Conduction Cooldown Test Facility (CCCTF), the Vertical Counting System (VCS), and the Irradiated Microsphere Gamma Analyzer (IMGA). The CCCTF includes liquid-nitrogen-cooled traps to extract 85 Kr out of the He sweep gas that passes through the furnace in which the fuel compacts are heated during safety testing. Analysis of the 85 Kr activity in the traps is the primary indicator for TRISO failure during safety testing. The VCS is a system used to accurately measure gamma emission from components placed in a lead-shielded chamber. It is used to count the CCCTF deposition cups after removal from furnace. Each cup resides in the CCCTF furnace for typically 12–24 h and is periodically replaced with a fresh cup throughout the safety test. Metallic fission products collect on the water-cooled cups and several gamma-emitting isotopes ( 110 mAg, 134 Cs, 137 Cs, 154 Eu, and 155 Eu) are often measured and provide indication of the retention performance of the TRISO coatings. The VCS is also used to measure the presence of these isotopes on the CCCTF tantalum liner and sweep gas inlet tube for the determination of cup collection efficiency, as well as support other gamma spectrometry needs related to calibration of the 85 Kr fission gas traps and various other special PIE tasks. The IMGA uses gamma spectrometry to measure the inventory of gamma-emitting isotopes in individual TRISO particles. An automated particle handling system within the IMGA hot cell removes each particle from a source vial and positions it in front of a gamma detector, and output from the gamma spectrometer is used by the IMGA software to determine a destination vial such that particles are sorted according to their inventory and retention characteristics. At the conclusion of the AGR-1 and AGR-2 PIE campaigns, the gamma spectrometer systems used at ORNL to support that PIE had reached the end of its life cycle due to gradual obsolescence of the hardware and software. Upgrade of the Canberra Genie 2000 software used by these systems to a Windows 10 version was not a viable option, because the newest Windows 10 version offered by Mirion (the new owner of the Canberra technology) did not include the dynamic-link libraries (DLLs) needed for integration with the custom PIE software used with the CCCTF and IMGA, and Mirion had no current plans for development and release of Windows 10 versions of these DLLs with the Model S560 Genie 2000 Programming Library. Ultimately a switch was made to ORTEC gamma spectrometry systems, which appeared to be a more sustainable solution due to more proactive vendor support. The ORTEC conversion involved replacing the aging detector preamplifier and multichannel analyzer (MCA) hardware, upgrading the obsolete Windows 7 computers to Windows 10 compatible models, adopting ORTEC GammaVision software, and extensive modification of the ORNL-developed Visual Basic .NET (VB.NET) programs that provide the CCCTF and IMGA user interfaces.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2023

Advective gas flow in bentonite: Development and comparison of enhanced multi-phase numerical approaches

Understanding the impact of repository gas, generated from degradation of waste and its interaction with the host rock, is essential when assessing the performance and safety function of long-term disposal systems for radioactive waste. Numerical models based on conventional multi-phase flow theory have historically been applied to predict the outcome and impact of gas flow on different repository components. However, they remain unable to describe the full complexity of the physical processes observed in water-saturated experiments (e.g., creation of dilatant pathways) and thus, the development of novel representations for their description is required when assessing fully saturated clay-based systems. This was the primary focus of Task A within the international cooperative project DECOVALEX-2019 (D-2019) and refinement of these approaches is the primary focus of this study (Task B in the current phase of DECOVALEX-2023). This paper summarises development of enhanced numerical representations of key processes and compares the performance of each model against high-quality laboratory test data. Experimental data reveals that gas percolation in water-saturated compacted bentonite is characterised by four key features: (i) a quiescence phase, followed by (ii) the gas breakthrough, which leads to a (iii) peak value, which is then followed by (iv) a negative decay. Three models based on the multiphase flow theory have been developed. These models can provide good initial values and reasonable responses for gas breakthrough (although some of them still predict a too-smooth response). Peak gas pressure values are in general reasonably well captured, although maximum radial stress differences are observed at 48 mm from the base of the sample. Here, numerical peak values of 12.8 MPa are predicted, whereas experimental values are about 11 MPa. These models are also capable of providing a reasonable representation of the negative pressure decay following peak pressure. However, other key specific features (such as the timing of gas breakthrough) still require a better representation. The model simulations and their comparison with experimental data show that these models need to be further improved with respect to model parameter calibration, the numerical representation of spatial heterogeneities in material properties and flow localisation, and the upscaling of the related physical processes and parameters. To further understand gas flow localisation, a new conceptual model has been developed, which shows that discrete channels can possibly be induced through the instability of gas-bentonite interface during gas injection, thus providing a new perspective for modeling gas percolation in low-permeability deformable media.

58 GEOSCIENCES↗

DOE OSTI2026

The Myth of Fungible FTE: A Quantitative Assessment of Matrixed Resource Allocation

Matrix organizations allow scientific facilities to share specialized personnel across projects, operations, maintenance, and strategic initiatives. Nominal staffing allocations, however, may not capture the schedule consequences of fragmented individual commitments, limited access to specialist groups, and intermittent availability of key decision makers. We developed a stochas- tic, daily-time-step simulation of a hypothetical medium-sized accelerator-facility project com- prising sequential phases and parallel tasks. Each task requires role-specific work measured in FTE-days. Ordinary personnel may be unavailable because they contribute concurrently to other institutional activities, while designated key roles have independently specified daily un- availability probabilities. An organization-wide priority factor scales the number of people from each functional group who can effectively contribute to the project. It is interpreted as a composite proxy for project access and workforce fragmentation across competing commit- ments. We examined project completion time as a function of this factor and Project Lead unavailability using 100 Monte Carlo runs per condition. Increasing priority factor from 0.1 to 1.0 reduced median completion time from 1708.5 days (interquartile range 1681.5–1735.25) to 390 days (interquartile range 379–399). At priority factor = 0.1, increasing Project Lead unavailability from 0.5 to 0.9 increased median completion time from 1713.5 days (interquartile range 1691–1733.25) to 4,417 days (interquartile range 4271.75–4550.5). The model quantifies the commonly expected sensitivity of project schedules to fragmented resource commitments and limited coordination availability. Within this model, the results also indicate a possible threshold regime in which small increases in workforce availability yield only modest sched- ule improvements until sufficient capacity becomes accessible, after which project performance improves sharply. With further validation and calibration, this quantitative framework could support resource-allocation decisions during initial project planning and subsequent schedule rebaselining.

Bai, Mei [SLAC National Accelerator Laboratory (SL↗

DOE OSTI2021

When and why PINNs fail to train: A neural tangent kernel perspective

Physics-informed neural networks (PINNs) have lately received great attention thanks to their flexibility in tackling a wide range of forward and inverse problems involving partial differential equations. However, despite their noticeable empirical success, little is known about how such constrained neural networks behave during their training via gradient descent. More importantly, even less is known about why such models sometimes fail to train at all. Here in this work, we aim to investigate these questions through the lens of the Neural Tangent Kernel (NTK); a kernel that captures the behavior of fully-connected neural networks in the infinite width limit during training via gradient descent. Specifically, we derive the NTK of PINNs and prove that, under appropriate conditions, it converges to a deterministic kernel that stays constant during training in the infinite-width limit. This allows us to analyze the training dynamics of PINNs through the lens of their limiting NTK and find a remarkable discrepancy in the convergence rate of the different loss components contributing to the total training error. To address this fundamental pathology, we propose a novel gradient descent algorithm that utilizes the eigenvalues of the NTK to adaptively calibrate the convergence rate of the total training error. Finally, we perform a series of numerical experiments to verify the correctness of our theory and the practical effectiveness of the proposed algorithms.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

NASA NTRSDate not supplied

The Lunar GNSS Receiver Experiment (LuGRE)

The Lunar GNSS Receiver Experiment (LuGRE) is a joint NASA-Italian Space Agency (ASI) payload on the Firefly Blue Ghost Mission 1 (BGM1) with the goal to demonstrate GNSS-based positioning, navigation, and timing at the Moon. LuGRE was chosen by the NASA Commercial Lunar Payload Services (CLPS) program as one of ten payloads on its “19D” task order for delivery to the lunar surface in 2023. The LuGRE payload consists of a weak-signal GNSS receiver, a high-gain L-band patch antenna, a low-noise amplifier, and an RF filter. The receiver will track GPS L1 C/A and L5, and Galileo E1 and E5a signals and will return pseudorange, carrier phase, and Doppler measurements to the ground. It will also calculate least-squares point solutions and Kalman-filter based navigation solutions onboard. In addition, the receiver features the capability to record raw I/Q baseband samples for downlink and ground processing. LuGRE will build on the legacy of prior missions in the Space Service Volume (SSV) including the initial experiments by AMSAT-OSCAR 40 and others, the GOES-R series of geostationary weather satellites, and the NASA Magnetospheric Multiscale (MMS) mission currently operating on GPS-based navigation at nearly 50% of lunar distance. Further, LuGRE will be one of the very first demonstrations of GNSS signal reception and navigation in the lunar environment and on the lunar surface, paving the way for operational use by future lunar missions such as Orion, Gateway, robotic and human landers, and surface rovers. Ultimately, all LuGRE science data will be released to a public data archive for the benefit of the GNSS and space communities. This paper provides a detailed overview of the LuGRE payload, including its design, concept of operations, and its predicted ability to meet its core science objectives. The baseline science investigations and priorities are outlined. Simulated performance results are shown based on the latest calibrated models including signal strength, signal availability, onboard navigation performance and convergence properties, and ground-based post-processed navigation performance.

LuGRE↗

NASA NTRSDate not supplied

Hybrid Model Based Approaches for Systems Health Management and Prognostics

To facilitate and solve the prediction problem, awareness of the current health state of the system is key, since it is necessary to perform condition-based predictions. To accurately predict the future state of any system, it is required to possess knowledge of its current health state and future operational conditions. Latest achievements of data-driven algorithms in regression of complex nonlinear functions and classification tasks have generated a growing interest in artificial intelligence for industrial applications. Complex multi-physics models as well as digital twins, once purely built on physics and corresponding simplified lumped parameter iterations, can now benefit from machine learning algorithms to mitigate the lack of understanding of some complex behavior. Given models of the current and future system behavior, a general approach of model-based prognostics can solve the prediction problem and further decision making. In principle, data driven approaches can replace expensive experimental test-setups as well as reduce the number of simulations needed to explore, e.g., the parametric space of a multi-parameter model. Nonetheless, the limitations of pure data-driven methods came to light rather quickly, at least for some industries. In many industrial applications, data acquisition is costly, and the volume of data that can be collected does not satisfy the requirements for an effective model training and cross-validation. Therefore, some recent works in the area of machine learning is focusing on blending physics with data-driven algorithms, thus mitigating the drawbacks of the two approaches and emphasizing respective advantages. Partial physical knowledge of the problem can aid the learning process by “guiding” the algorithm towards efficient solutions that satisfy the physics driving the system behavior. The result is a hybrid modeling approach combining physical knowledge as well data driven methods to develop a unified hybrid approach. A hybrid framework for fusing information from physics-based performance models along with deep learning algorithms for prognostics of complex safety critical systems is presented. In this framework, physics-based performance models infer unobservable model parameters related to the system's components health solving a calibration problem in the deep learning approach.

Hybrid Modeling↗

NASA NTRSDate not supplied

Hybrid Approaches to Systems Health Management and Prognostics

To facilitate and solve the prediction problem, awareness of the current health state of the system is key, since it is necessary to perform condition-based predictions. To accurately predict the future state of any system, it is required to possess knowledge of its current health state and future operational conditions. Latest achievements of data-driven algorithms in regression of complex nonlinear functions and classification tasks have generated a growing interest in artificial intelligence for industrial applications. Complex multi-physics models as well as digital twins, once purely built on physics and corresponding simplified lumped parameter iterations, can now benefit from machine learning algorithms to mitigate the lack of understanding of some complex behavior. Given models of the current and future system behavior, a general approach of model-based prognostics can solve the prediction problem and further decision making. In principle, data driven approaches can replace expensive experimental test-setups as well as reduce the number of simulations needed to explore, e.g., the parametric space of a multi-parameter model. Nonetheless, the limitations of pure data-driven methods came to light rather quickly, at least for some industries. In many industrial applications, data acquisition is costly, and the volume of data that can be collected does not satisfy the requirements for an effective model training and cross-validation. Therefore, some recent works in the area of machine learning is focusing on blending physics with data-driven algorithms, thus mitigating the drawbacks of the two approaches and emphasizing respective advantages. Partial physical knowledge of the problem can aid the learning process by “guiding” the algorithm towards efficient solutions that satisfy the physics driving the system behavior. The result is a hybrid modeling approach combining physical knowledge as well data driven methods to develop a unified hybrid approach. A hybrid framework for fusing information from physics-based performance models along with deep learning algorithms for prognostics of complex safety critical systems is presented. In this framework, physics-based performance models infer unobservable model parameters related to the system's components health solving a calibration problem in the deep learning approach.

Systems Health Management↗