Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Reliability Specifications”

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
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

MOD-Amp System Design Spring: Spring 2026 – Georgetown University, SYSM–5620

High-energy lasers (HELs) play an important role in both national defense and scientific research. In defense applications, HELs are used for target detection, tracking, and engagement. In research environments, they support studies of extreme physical conditions relevant to fusion energy and plasma science. These systems depend on the amplification of light through stimulated emission of radiation, allowing optical energy to be increased to the levels required for operation. This amplification occurs when light passes through an energized gain medium that receives energy from an external optical or electrical source. To achieve the desired output, laser systems often use multiple amplification stages, including high-gain preamplifiers and lower-gain power or booster amplifiers. At Lawrence Livermore National Laboratory (LLNL) and other national laboratories, many large-aperture laser amplifier systems are aging and rely on system-specific hardware, obsolete technologies, and incomplete documentation. These legacy systems create challenges for maintenance, supportability, and long-term operation. Their lack of standardization also increases the difficulty of sustaining reliable performance over time. As this infrastructure continues to age, the likelihood of unplanned downtime grows, which can negatively affect both national security missions and scientific research programs that depend on dependable HEL capabilities. The purpose of this document is to demonstrate the application of systems engineering fundamentals and design thinking through the development of a laser amplifier case study. The proposed system concept is intended as an academic exercise and not as a finalized engineering design. As a result, the development presented in this document is incomplete and may contain technical assumptions or errors that would require further investigation before any real-world implementation.

42 ENGINEERING↗
DOE OSTI2022

Tracing and Forecasting Metabolic Indices of Cancer Patients Using Patient-Specific Deep Learning Models

We develop a patient-specific dynamical system model from the time series data of the cancer patient’s metabolic panel taken during the period of cancer treatment and recovery. The model consists of a pair of stacked long short-term memory (LSTM) recurrent neural networks and a fully connected neural network in each unit. It is intended to be used by physicians to trace back and look forward at the patient’s metabolic indices, to identify potential adverse events, and to make short-term predictions. When the model is used in making short-term predictions, the relative error in every index is less than 10% in the L ∞ norm and less than 6.3% in the L 1 norm in the validation process. Once a master model is built, the patient-specific model can be calibrated through transfer learning. As an example, we obtain patient-specific models for four more cancer patients through transfer learning, which all exhibit reduced training time and a comparable level of accuracy. This study demonstrates that this modeling approach is reliable and can deliver clinically acceptable physiological models for tracking and forecasting patients’ metabolic indices.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2025

Eucalyptus – An Analysis Suite for Fault Trees with Uncertainty Quantification

Eucalyptus is a novel code developed at Lawrence Livermore National Laboratory to incorporate uncertainty quantification into Fault Tree Analysis (FTA). This tool addresses the challenge of imperfect knowledge in “grey-box” systems by allowing analysts to incorporate and propagate uncertainty from component-level assessments to system-level effects. Eucalyptus facilitates a consistent evaluation of the impact of subject matter expert judgment and knowledge gaps on overall system response by Monte Carlo generation of possible system fault trees, sampling probabilities of the existence of subsystems and components. Here, the code supports the specification of fault trees through text and allows export to various formats, including auto-generated images, easing analysis and reducing errors. It has undergone extensive verification testing, demonstrating its reliability and readiness for deployment, and leverages on-node parallelism for rapid analysis. Example analyses are shown that include the identification of system failure paths and quantification of the value of further information about system components.

Fault Tree Analysis↗
DOE OSTI2024

Next-Generation, High-temperature, High-frequency, High-efficiency, High-power-density Traction System

To meet performance and reliability requirements necessary for broader adoption of electric drive vehicles, the Electrical and Electronics Technical Team of the U.S. Drive partnership has established aggressive design goals for next-generation electric vehicle drivetrains. Specifically, the 2025 roadmap stipulates a 100 kW/L power density target and a $\$$2.7/kW cost target for power electronics, in addition to high-voltage operation (i.e., greater than 800 VDC). The additional targets for traction motor and the overall system performance impose further challenges on the power electronics design. For example, many high specific power machines have reduced iron content, and therefore reduced intrinsic filtering, thus requiring the inverter to supply a low-distortion drive current. These machines also typically have a high pole count, thus requiring drive current at a higher electrical frequency. Other motors, such as brush-less dc and switch reluctance machines, require a carefully-shaped, non-sinusoidal drive current (Yang, Shang, Brown, & Krishnamurthy, 2015), (Zhang, Bowman, O'Connel, & Haran, 2018), (Anderson, et al., 2018). Two- and three-level inverter topologies are the conventional framework for the power electronics design of the drivetrain, and some demonstrations have shown recent progress towards addressing cost, power density and efficiency goals (Gurpinar & Ozpineci, 2018), (Zhu, Kim, Chen, Erickson, & Maksimović, 2018), (Deshpande, Chen, Narayanasamy, Sathyanarayanan, & Luo, 2018), (Alizadeh, et al., 2019). However, an unconventional approach may be necessary to take the dramatic leap in power density necessitated by the roadmap—while simultaneously addressing the other system needs. Therefore, this project leverages the flying capacitor multilevel (FCML) topology, together with a scalable, modular approach, to address these needs. This type of hybrid converter has several advantages: lower voltage (i.e., less than 300 V) transistors can be used, energy-dense capacitors process most of the power, and the output current waveform is multilevel and exhibits a frequency multiplying effect—in other words, the output has reduced dv/dt and filtering requirements for the same high voltage dc bus. For example, in an electric vehicle with an 800 V bus, a 10-level FCML could leverage 100 V, commercially available GaN devices switching at 115 kHz to produce a ~1 MHz switching waveform (modulated according to the motor drive requirements) with one ninth of the dv/dt of a two-level converter. Prior work has already demonstrated promising performance and gravimetric power density figures for more electric aircraft applications (Pallo, Foulkes, Modeer, Coday, & Pilawa-Podgurski, 2018). This project leverages lessons learned to achieve the volumetric power density of 100 kW/L by employing advanced liquid cooling, address the 300,000 mile reliability challenge with redundant design, topology failure studies and online health monitoring, and reduce costs to $\$$2.7/kW through the use of low-cost GaN devices, modular converter assemblies, and modest modifications to traditional manufacturing methods. The project involved several hardware designs, each achieving increasing performance. At the conclusion of the project, a volumetric power density of 380 kW/L was achieved, in a 800V dc-ac converter, greatly surpassing even the aggressive target goal.

33 ADVANCED PROPULSION SYSTEMS↗
DOE OSTI2025

TASTI-GRID: Overview of the State of Oregon’s Resilience and Reliability [Slides]

This report is intended to help your state identify the most effective investments to improve grid resilience and reliability, based on analysis of outage data, weather events, and the current state of the electric grid. The information presented is derived from the best available data. The discussions and recommendations included provide context and valuable insights into potential high returns on investment (ROI). Specifically, the report offers an overview of electric grid resilience in Oregon, covering outage information, reasons for outages, calculated "resilience scores" across the state, and recommendations for both immediate and long-term investments in grid infrastructure. The document provides a concise overview of the analytical capabilities of the TASTI-GRID team. By utilizing additional data and resources, it synthesizes insights from the platform to present a distinct perspective on investing in grid resilience throughout Oregon. This document is not intended to influence or direct state-level decisions related to investments in grid resilience and reliability. TASTI-GRID should not be used for real-time monitoring, supporting the ESF #12 emergency response functions, or predicting the restoration times for outages.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2022

CLASSY IV. Exploring UV Diagnostics of the Interstellar Medium in Local High- z Analogs at the Dawn of the JWST Era

The Cosmic Origins Spectrograph (COS) Legacy Archive Spectroscopic SurveY (CLASSY) provides the first high resolution spectral catalog of 45 local high-z analogs in the ultraviolet (UV; 1200–2000 Å) to investigate their stellar and gas properties. Here we present a toolkit of UV interstellar medium (ISM) diagnostics, analyzing the main emission lines of CLASSY spectra (N IV] λλ1483,87, C IV λλ1548,51, He IIλ1640, O III]λλ1661,6, Si III] λλ1883,92, C III] λ1907,9). Specifically, our aim is to provide accurate diagnostics for the reddening E(B - V), electron density n e , electron temperature T e , metallicity 12+log(O/H), and ionization parameter log(U), taking the different ISM ionization zones into account. We calibrate our UV toolkit using well-known optical diagnostics, analyzing archival optical spectra for all CLASSY targets. We find that UV density diagnostics estimate n e values that are ~1–2 dex higher (e.g., n e (C III] λλ1907,9) ~ 10 4 cm -3 ) than those inferred from their optical counterparts (e.g., n e ([S II]λλ6717,31) ~ 10 2 cm -3 ; n e ([Ar IV]λλ4714,41) ~ 10 3 cm -3 ). T e derived from the hybrid ratio [O III] λ1666/λ5007 proves to be reliable, implying differences in determining 12+log(O/H) compared to the optical counterpart O III] λ4363/[O III] λ5007 within ~ ±0.3 dex. We also investigate the relation between the stellar and gas E(B - V), finding consistent values at high specific star formation rates (sSFRs; log sSFR ≳ -8 yr -1 ), while at low sSFRs we confirmed an excess of dust attenuation in the gas. Finally, we investigate UV line ratios and equivalent widths to provide correlations with 12+log (O/H) and log(U), but note that there are degeneracies between the two. With this suite of UV-based diagnostics, we illustrate the pivotal role CLASSY plays in understanding the chemical and physical properties of high-z systems that JWST can observe in the rest-frame UV.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

Evaluation of Saccadic Component Measure on Smooth Pursuit Tests

ABSTRACT Introduction Despite the advancement of eye-tracking technology for smooth pursuit (SP) eye movement evaluation, qualitative observation offers much information that is not captured by computers; hence, both objective and qualitative information should be utilized to evaluate SP. This study examined the consistency among our clinicians when evaluating SP using normal (N), grossly normal (GN), mildly abnormal (MA), and abnormal (AB) as classifications. We then evaluated the effect of combining GN and MA into a single subclinical (SUBC) category. We also evaluated the computerized percent saccade (PS) metric by determining its sensitivity and specificity in classifying SP. Materials and Methods Retrospective horizontal and vertical SP test videos and numerical data for 70 participants were obtained from the Neuro Kinetics Neuro-Otologic Test Center and de-identified. From this, eye-tracking videos, time plots of eye-tracking positional data, and tables of SP eye-tracking performance data were generated for 0.1, 0.3, and 0.5 Hz in both horizontal and vertical planes, totaling 6 tests per subject. Three clinicians rated each subject’s SP performance as N, GN, MA, or AB for a total of 6 ratings (3 frequencies, horizontal and vertical). This process was repeated using N, SUBC, and AB as rating categories. Clinicians also provided an overall SP rating for each plane as follows: AB if the results were abnormal for 2 or more frequencies tested. Alternatively, if fewer than 2 frequencies presented with a rating of AB, then an overall rating of MA, GN, or N was determined at the respective clinician’s discretion. Results When the 3 clinicians were tasked with classifying SP videos using 4 clinical categories, fair overall agreement was demonstrated. However, when MA and GN categories were combined into an SUBC category, the overall agreement for the 3 clinicians improved slightly for both horizontal SP (HSP) and vertical SP (VSP). This pattern of agreement did not differ considerably when comparing HSP versus VSP, and good consistency and reliability was observed across clinicians. Again, inter-rater consistency was smaller for VSP versus HSP despite the reduction in clinical categories. Cut-off values were generated for the PS metric and demonstrated good specificity and sensitivity when they were exceeded for 2 or more frequencies in a particular plane when evaluating a subject’s SP test. Conclusions

General & Internal Medicine↗
DOE OSTI2023

A Hybrid Physics-Based, Data-Driven Approach to Model Damage Accumulation in Corrosion of Polymeric Adhesives

The degradation of polymeric adhesives in dissimilar joints is a menace to electric and structural components of the automotive industry. In polymeric adhesives, failure often occurs due to damage accumulated from discrete environmental sources—thermal degradation, oxidation, hydrolysis, radiation, and moisture—that are combined with progressive damage mechanisms, such as aging and fatigue. Because most damage mechanisms that lead to degradation act in parallel, the estimation of the lifetime of a polymeric material becomes a significant reliability issue, no study currently exists that can simultaneously consider even two of these phenomena. Most current efforts are directed toward a single-phenomenon, single-model approach in which a model is developed to cover one specific damage mechanism. While this approach is advantageous in exploring single mechanisms, it is strongly inconclusive for degradation that occurs by multiple damage mechanisms. Our team plans to take a fresh perspective by designing a hybrid constitutive/data-driven approach that allows models of different damage mechanisms to be integrated into one model and then used to explore the degradation-induced failure of polymeric adhesives. The theoretical-computational part of this work is divided into four research components as described in the Approach section below.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Interrelationships among methods of estimating microbial biomass across multiple soil orders and biomes

Understanding the role of soil microbes is critical to ecosystem processes, and more thorough comparisons of measurement proxies for soil microbial biomass could broaden the inclusion of explicit microbial parameterization in soil carbon cycling and earth system models. We measured physical, chemical, and biological data from eight soil orders representing 11 major biomes and four climate regions. Four prominent methods to measure microbial abundance—chloroform fumigation extraction (CFE), total DNA yield, gene copy number by quantitative polymerase chain reaction (GCN), and phospholipid fatty acids (PLFA)—were compared to assess their relationships with each other and with soil characteristics. Correlations were observed when comparing methods, with CFE correlating strongly with total DNA yield, GCN, and PLFA; CFE with bacterial GCN and bacterial PLFA; and to a lesser extent, total PLFA and total DNA yield. Correlations improved with the removal of organic soils (Histosols, Gelisols). Comparisons involving extracted DNA were improved by correcting for clay content, due to DNA extraction inefficiencies in clay-rich soils. Correlations involving fungi (PLFA or GCN) were always less significant. These methods could serve as reliable, inter-relatable proxies for the estimation of total soil microbial biomass while recognizing that the proxies are less effective at parsing differences between bacteria and fungi. Here, we provide specific equations to relate measures of soil microbial biomass by these four different methods to enable microbial models to utilize a greater diversity of observed data sources in parameterizations and simulations. Caveats for the equations and their values are also discussed.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Increasing the Reproducibility and Replicability of Supervised AI/ML in the Earth Systems Science by Leveraging Social Science Methods

Artificial intelligence (AI) and machine learning (ML) pose a challenge for achieving science that is both reproducible and replicable. The challenge is compounded in supervised models that depend on manually labeled training data, as they introduce additional decision-making and processes that require thorough documentation and reporting. We address these limitations by providing an approach to hand labeling training data for supervised ML that integrates quantitative content analysis (QCA)—a method from social science research. The QCA approach provides a rigorous and well-documented hand labeling procedure to improve the replicability and reproducibility of supervised ML applications in Earth systems science (ESS), as well as the ability to evaluate them. Specifically, the approach requires (a) the articulation and documentation of the exact decision-making process used for assigning hand labels in a “codebook” and (b) an empirical evaluation of the reliability” of the hand labelers. In this paper, we outline the contributions of QCA to the field, along with an overview of the general approach. We then provide a case study to further demonstrate how this framework has and can be applied when developing supervised ML models for applications in ESS. With this approach, we provide an actionable path forward for addressing ethical considerations and goals outlined by recent AGU work on ML ethics in ESS.

58 GEOSCIENCES↗
DOE OSTI2025

Navigating the Noise: Bringing Clarity to ML Parameterization Design With O $\boldsymbol{\mathcal{O}}$(100) Ensembles

Abstract Machine‐learning (ML) parameterizations of subgrid processes (here of turbulence, convection, and radiation) may one day replace conventional parameterizations by emulating high‐resolution physics without the cost of explicit simulation. However, uncertainty about the relationship between offline and online performance (i.e., when integrated with a large‐scale general circulation model) hinders their development. Much of this uncertainty stems from limited sampling of the noisy, emergent effects of upstream ML design decisions on downstream online hybrid simulation. Our work rectifies the sampling issue via the construction of a semi‐automated, end‐to‐end pipeline for size ensembles of hybrid simulations, revealing important nuances in how systematic reductions in offline error manifest in changes to online error and online stability. For example, removing dropout and switching from a Mean Squared Error to a Mean Absolute Error loss both reduce offline error, but they have opposite effects on online error and online stability. Other design decisions, like incorporating memory, converting moisture input from specific humidity to relative humidity, using batch normalization, and training on multiple climates do not come with any such compromises. Finally, we show that ensemble sizes of may be necessary to reliably detect causally relevant differences online. By enabling rapid online experimentation at scale, we can empirically settle debates regarding subgrid ML parameterization design that would have otherwise remained unresolved in the noise.

Lin, Jerry [Department of Earth System Sciences Un↗
DOE OSTI2020

First-principles modeling of chemistry in mixed solvents: Where to go from here?

Mixed solvents (i.e., binary or higher order mixtures of ionic or nonionic liquids) play crucial roles in chemical syntheses, separations, and electrochemical devices because they can be tuned for specific reactions and applications. Apart from fully explicit solvation treatments that can be difficult to parameterize or computationally expensive, there is currently no well-established first-principles regimen for reliably modeling atomic-scale chemistry in mixed solvent environments. We offer our perspective on how this process could be achieved in the near future as mixed solvent systems become more explored using theoretical and computational chemistry. We first outline what makes mixed solvent systems far more complex compared to single-component solvents. An overview of current and promising techniques for modeling mixed solvent environments is provided. We focus on so-called hybrid solvation treatments such as the conductor-like screening model for real solvents and the reference interaction site model, which are far less computationally demanding than explicit simulations. We also propose that cluster-continuum approaches rooted in physically rigorous quasi-chemical theory provide a robust, yet practical, route for studying chemical processes in mixed solvents.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

A foundation model for atomistic materials chemistry

Atomistic simulations of matter, especially those that leverage first-principles (ab initio) electronic structure theory, provide a microscopic view of the world, underpinning much of our understanding of chemistry and materials science. Over the last decade or so, machine-learned force fields have transformed atomistic modeling by enabling simulations of ab initio quality over unprecedented time and length scales. However, early machine-learning (ML) force fields have largely been limited by (i) the substantial computational and human effort required to develop and validate potentials for each particular system of interest and (ii) a general lack of transferability from one chemical system to the next. Here, we show that it is possible to create a general-purpose atomistic ML model, trained on a public dataset of moderate size, that is capable of running stable molecular dynamics for a wide range of molecules and materials. We demonstrate the power of the MACE-MP-0 model-and its qualitative and at times quantitative accuracy-on a diverse set of problems in the physical sciences, including properties of solids, liquids, gases, chemical reactions, interfaces, and even the dynamics of a small protein. The model can be applied out of the box as a starting or "foundation" model for any atomistic system of interest and, when desired, can be fine-tuned on just a handful of application-specific data points to reach ab initio accuracy. Establishing that a stable force-field model can cover almost all materials changes atomistic modeling in a fundamental way: experienced users obtain reliable results much faster, and beginners face a lower barrier to entry. Foundation models thus represent a step toward democratizing the revolution in atomic-scale modeling that has been brought about by ML force fields.

Batatia, Ilyes↗
DOE OSTI2025

LLaMP v0.1.0

Reducing hallucination of Large Language Models (LLMs) is imperative for use in the sciences, where reliability and reproducibility are crucial. However, LLMs inherently lack long-term memory, making it a nontrivial, ad hoc, and inevitably biased task to fine-tune them on domain-specific literature and data. LLaMP is a multimodal retrieval-augmented generation (RAG) framework of hierarchical reasoning and acting (ReAct) agents that can dynamically and recursively interact with Materials Project to ground large language models on high-fidelity materials informatics.

Riebesell, Janosh [Lawrence Berkeley National Labo↗
DOE OSTI2023

Improving Subsurface Stress Characterization for Carbon Dioxide Storage Projects by Incorporating Machine Learning Techniques

The overall objective of this project is to develop a framework for reliable characterization and prediction of the state of stress in the overburden and underburden (including the basement) in CO 2 storage reservoirs using machine learning and integrated geomechanics and geophysical methods. Specifically, we propose to develop workflow encompassing of technologies and/or methods to predict stress and pressure changes due to CO 2 injection in an active tertiary recovery site and their impacts on subtle fault activation, fractures and occurrence of microseismic events and compare responses to field observations. In this project, we anticipate using dataset from the Farnsworth field Unit (FWU) which is operated by Purdure Petroleum. A novel elastic-waveform VSP inversion technique will be used to estimate high-resolution spatial and temporal changes of elastic moduli in CO 2 storage reservoirs, which will be combined with velocity-stress relationship derived from laboratory tests to obtain subsurface pressure and stress. Clustered microseismic data will be jointly inverted for improved focal mechanisms. Least-squares reverse-time migration of microseismic waveform data will be performed to directly image fracture/fault zones. Additionally, a deep neural network machine learning technique with convolutional and recurrent layers will be used for learning the spectro-temporal structures in microseismic waveforms. The results of this geotechnical data analysis will be integrated to develop a high-resolution 3D mechanical earth model extending from the overburden sealing formations to the underburden including the basement. Mechanical properties will be derived through integration of mechanical logs, tests, available results from chemo-mechanical laboratory tests, and elastic inversion of seismic data using a combination of Bayesian and stochastic methods as well as machine learning technique. Failure features (faults/fractures) will be represented and/or modeled based on seismic and core data analysis. A transient hydrodynamic-geomechanical model will be developed through coupling with the calibrated FWU reservoir simulation model. The full physics coupled model will be used to train a reduced order proxy model using machine learning algorithm for estimating stress which will then be used with appropriate constitutive relationships and forward seismological models to simulate pressure changes and induced microseismicity. An advanced optimization framework will be developed to perform a history match to minimize error between field observations and simulated. The history matched proxy model will be verified against the full-physics equivalent. The field observations that will be used in the coupled model calibration process include pressure/stress inverted from VSP, moment magnitude from microseismic analysis, real time downhole pressure measurements, production and injection data. Parameter sensitivity and uncertainty analysis will be performed to characterize the impact of model parameter uncertainty on stress estimates. The proposed project will have significant impact on future field implementation of the proposed technology. Because the project field site is an ongoing CO 2 EOR development, the value of the new technology will be demonstrated in an operational context and evaluated as a viable risk mitigation strategy. Cost/benefit will be evaluated together with the various commercial incentives for CO 2 sequestration available to oil and gas operators. The extensive available dataset and ongoing data acquisition under the SWP Phase III work plan provides flexibility for investigation of multiple approaches and reduces technical risk.

58 GEOSCIENCES↗
DOE OSTI2022

Always look on the bright side of life: Taking a different perspective on cybersecurity performance

This presentation will introduce participants to key mental models on how to think about and understand safety from two of the leading voices in the discipline – Dr. Erik Hollnagel and Dr. Todd Conklin – and from there we ask the simple question of “how can we apply this to cybersecurity?” At the completion of this presentation attendees will be able to describe Dr. Hollnagel’s Safety-II approach and Dr. Conkiln’s definition of safety as “not the absence of accidents but the presence of effective defenses.” They will be able to explain how these ideas translate to the cybersecurity world broadly, and more specifically to operational technology cybersecurity where physical safety consequences are coupled to control system performance. Finally, participants will be able to identify how selected principles from Weick and Sutcliffe’s high reliability organization theory can help operationalize these key mental models around safety, for improved organizational cybersecurity over the long run.

99 GENERAL AND MISCELLANEOUS↗
DOE OSTI2021

Surrogate Modeling For Efficiently Accurately and Conservatively Estimating Measures of Risk

We present a surrogate modeling framework for conservatively estimating measures of risk from limited realizations of an expensive physical experiment or computational simulation. We adopt a probabilistic description of risk that assigns probabilities to consequences associated with an event and use risk measures, which combine objective evidence with the subjective values of decision makers, to quantify anticipated outcomes. Given a set of samples, we construct a surrogate model that produces estimates of risk measures that are always greater than their empirical estimates obtained from the training data. These surrogate models not only limit over-confidence in reliability and safety assessments, but produce estimates of risk measures that converge much faster to the true value than purely sample-based estimates. We first detail the construction of conservative surrogate models that can be tailored to the specific risk preferences of the stakeholder and then present an approach, based upon stochastic orders, for constructing surrogate models that are conservative with respect to families of risk measures. The surrogate models introduce a bias that allows them to conservatively estimate the target risk measures. We provide theoretical results that show that this bias decays at the same rate as the L 2 error in the surrogate model. Our numerical examples confirm that risk-aware surrogate models do indeed over-estimate the target risk measures while converging at the expected rate.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Reliable and Efficient Machine Learning (Final Technical Report)

Modern scientific experiments generate massive amounts of data at a pace much faster than humans can manually analyze. While machine learning has revolutionized commercial data analysis (such as recommending movies or recognizing faces), applying these tools to complex scientific discovery is challenging because scientific answers must be precise, interpretable, and adhere to physical laws. The research under this project aims to develop new mathematical tools and computer algorithms specifically designed for scientific applications. Major progress has been made in automatically cleaning and deconstructing messy experimental data, analyzing the visual information of physical phenomena, determining the underlying physical variables, and providing rig orous mathematical analysis of interesting algorithms and concepts widely used in machine learning. This project addressed the critical gap between our ability to generate massive scientific data and our ability to extract interpretable information from it. We established mathematical foundations for Scientific Machine Learning (SciML) aimed at effective data analytics and automated discovery. Our work focused on three core objectives: (1) developing reliable feature extraction methods for dynamic high-dimensional data, (2) establishing mathematical foundations for discovering dynamics via neural networks, and (3) creating rigorous optimization techniques for these models. Key outcomes come from two fronts. On the practical side, they include the development of algorithms that significantly enhance the extraction of signals from field data, as well as the capability to handle situations that exhibit smooth variations or physical stretching due to temperature changes. They also include the creation of an automated framework for discovering fundamental state variables from raw experimental data, demonstrating the ability to identify intrinsic physical dimensions without prior knowledge of the governing laws. On the theoretical front, the research results in theoretical advances in Optimal Transport, a widely used notion in SciML, specifically regarding functions with fixed-size nodal sets, provide sharp bounds relevant to uncertainty quantification. Meanwhile, the outcomes also include the establishment of convergence theories for nonlocal gradient descent methods, enabling robust optimization with noisy data in high-dimensional settings commonly encountered in scientific modeling. The project also helps creating opportunities to train the next generation of researchers, equipping them with the necessary technical skills for today’s workplace and preparing them for future advances.

97 MATHEMATICS AND COMPUTING↗