Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Active Learning”

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 613 records · Page 34

Adaptive Interface-PINNs (AdaI-PINNs): An Efficient Physics-Informed Neural Networks Framework for Interface Problems

Here, we present an efficient physics-informed neural networks (PINNs) framework, termed Adaptive Interface-PINNs (AdaI-PINNs), to improve the modeling of interface problems with discontinuous coefficients and/or interfacial jumps. This framework is an enhanced version of its predecessor, Interface PINNs or I-PINNs (Sarma et al.; https://doi.org/10.1016/j.cma.2024.117135), which involves domain decomposition and assignment of different predefined activation functions to the neural networks in each subdomain across a sharp interface, while keeping all other parameters of the neural networks identical. In AdaI-PINNs, the activation functions vary solely in their slopes, which are trained along with the other parameters of the neural networks. This makes the AdaI-PINNs framework fully automated without requiring preset activation functions. Comparative studies on one-dimensional, two-dimensional, and three-dimensional benchmark elliptic interface problems reveal that AdaI-PINNs outperform I-PINNs, reducing computational costs by 2-6 times while producing similar or better accuracy.

97 MATHEMATICS AND COMPUTING↗

Predicting the Activity and Selectivity of Bimetallic Metal Catalysts for Ethanol Reforming using Machine Learning

Machine learning is ideally suited for the pattern detection in large uniform datasets, but consistent experimental datasets on catalyst studies are often small. Here we demonstrate how a combination of machine learning and first-principles calculations can be used to extract knowledge from a relatively small set of experimental data. The approach is based on combining a complex machine-learning model trained on an extensive computational library of transition-state energies with simple linear regression models of experimental catalytic activities and selectivities from the literature. Using the combined model, we identify the key C–C bond scission reactions involved in ethanol reforming and perform a computational screening for ethanol reforming on monolayer bimetallic catalysts with architectures TM-Pt-Pt(111) and Pt-TM-Pt(111) (TM = 3d transition metals). The model also predicts four promising catalyst compositions for future experimental studies. In conclusion, the approach is not limited to ethanol reforming but is of general use for the interpretation of experimental observations as well as for the computational discovery of novel catalytic materials.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Conditioned place preference and locomotor activity in response to methylphenidate, amphetamine and cocaine in mice lacking dopamine D4 receptors

Methylphenidate (MP) and amphetamine (AMPH) are the most frequently prescribed medications for the treatment of attention-deficit/hyperactivity disorder (ADHD). Both drugs are believed to derive their therapeutic benefit by virtue of their dopamine (DA)-enhancing effects, yet an explanation for the observation that some patients with ADHD respond well to one medication but not to the other remains elusive. The dopaminergic effects of MP and AMPH are also thought to underlie their reinforcing properties and ultimately their abuse. Polymorphisms in the human gene that codes for the DA D4 receptor (D4R) have been repeatedly associated with ADHD and may correlate with the therapeutic as well as the reinforcing effects of responses to these psychostimulant medications. Conditioned place preference (CPP) for MP, AMPH and cocaine were evaluated in wild-type (WT) mice and their genetically engineered littermates, congenic on the C57Bl/6J background, that completely lack D4Rs (knockout or KO). In addition, the locomotor activity in these mice during the conditioning phase of CPP was tested in the CPP chambers. D4 receptor KO and WT mice showed CPP and increased locomotor activity in response to each of the three psychostimulants tested. D4R differentially modulates the CPP responses to MP, AMPH and cocaine. While the D4R genotype affected CPP responses to MP (high dose only) and AMPH (low dose only) it had no effects on cocaine. Inasmuch as CPP is considered an indicator of sensitivity to reinforcing responses to drugs these data suggest a significant but limited role of D4Rs in modulating conditioning responses to MP and AMPH. In the locomotor test, D4 receptor KO mice displayed attenuated increases in AMPH-induced locomotor activity whereas responses to cocaine and MP did not differ. These results suggest distinct mechanisms for D4 receptor modulation of the reinforcing (perhaps via attenuating dopaminergic signalling) and locomotor properties of these stimulant drugs. Thus, individuals with D4 receptor polymorphisms might show enhanced reinforcing responses to MP and AMPH and attenuated locomotor response to AMPH.

59 BASIC BIOLOGICAL SCIENCES↗

Understanding EV Charging Pain Points Through Deep Learning Analysis

Current and potential electric vehicle (EV) owners express concerns about the charging infrastructure, mentioning non-functional chargers, prolonged charging times, inconvenient charger locations, long wait times, and high costs as major barriers. Addressing these issues often requires analyzing actual vehicle charging data, which is typically proprietary and inconsistent due to diverse standards and protocols. To understand and improve the EV charging experience, customer reviews are typically used to identify common customer pain points (CPPs). However, there is not a comprehensive method to map customer reviews to a standardized set of CPPs. In collaboration with the National Charging Experience (ChargeX) Consortium, this study bridges these gaps by proposing a Systematic Categorization and Analysis of Large-scale EV-charging Reviews (SCALER) framework. SCALER is an integrated, deep learning framework that segments, actively labels, analyzes, and classifies EV charging customer reviews into six CPP categories. To test its effectiveness, we used SCALER to analyze over 72,000 reviews from customers charging various EV models on different networks across the United States. SCALER achieves a classification accuracy of 92.5%, with an F1 score exceeding 85.7%. By demonstrating real-world applications of SCALER, we enhance the industry’s ability to understand and address CPPs to improve the EV charging experience.

29 - ENERGY PLANNING, POLICY AND ECONOMY↗

GLUE Code: A framework handling communication and interfaces between scales

Many scientific applications are inherently multiscale in nature. Such complex physical phenomena often require simultaneous execution and coordination of simulations spanning multiple time and length scales. This is possible by combining expensive small-scale simulations (such as molecular dynamics simulations) with larger scale simulations (such continuum limit/hydro solvers) to allow for considerably larger systems using task and data parallelism. However, the granularity of the tasks can be very large and often leads to load imbalance. Traditionally, we use approximations to streamline the computation of the more costly interactions and this introduces trade-offs between simulation cost and accuracy. In recent years, the available computational power and the advances in machine learning have made computing these scale-bridging interactions and multiscale simulations more feasible. One driving application has been plasma modeling in inertial confinement fusion (ICF), which is fundamentally multiscale in nature. This requires deep understanding of how to extrapolate microscopic information into macroscopically relevant scales. For example, in ICF one needs an accurate understanding of the connection between experimental observables and the underlying microphysics. The properties of the larger scales are often affected by the microscale behavior incorporated usually into the equations of state and ionic and electronic transport coefficients (Liboff, 1959; Rinderknecht et al., 2014; Rosenberg et al., 2015; Ross et al., 2017). Instead of incorporating this information using reliable molecular dynamics (MD) simulations, one often needs to use theoretical models, due to the inability of MD to reach engineering scales (Glosli et al., 2007; Marinak et al., 1998). One approach to resolve this issue is by coupling two MD simulations of different scales via force interpolation, e.g., the AdResS method (Krekeler et al., 2018; Nagarajan et al., 2013). Another approach, which we will pursue in the scope of this work, is by enabling scale bridging between MD simulations and meso/macro-scale models through the development and support of application programming interfaces that these different applications can interact through.

54 ENVIRONMENTAL SCIENCES↗

Identifying recharge sources and their impacts on a North Central New Mexico shallow aquifer using unsupervised machine learning

In this article, shallow aquifers are important but highly variable resources in arid to semi-arid regions. Limited shallow aquifer volume results in high sensitivity to recharge fluctuations, which can impact the local fauna and flora, and transport of contaminants in the aquifer or vadose zone. Aquifer response to external forcing (e.g., precipitation) is usually solved by estimating aquifer parameters and running physics-based models to match known fluctuations of hydraulic head. However, this technique is time and computationally expensive. Furthermore, high aquifer complexity decreases precision in physics-based models. Alternatively supervised machine learning is used to predict aquifer dynamics. However, these techniques rely on input data and struggle to interpret aquifer response for missing sources (i.e., snowpack data). To counter these problems, we propose an unsupervised machine learning technique (NMFk) to estimate the impact of different sources on aquifer recharge. NMFk is used to understand the influence of external forcing on shallow aquifer recharge in the Pajarito Plateau (Los Alamos, NM, USA). The results show how NMFk can be used to reduce the data dimension in a complex field dataset to three recharge signals that cause fluctuations within the field data. Here, the source signals are interpreted as rainfall, snowmelt, and a delayed aquifer response to the previous two signals. These results evidence how heterogeneous aquifers delimited by canyons incised into the Pajarito Plateau respond in similar ways to the source signals identified by NMFk. Furthermore, results show the importance of the local geology where faults act as sinks, and anthropogenic disturbances can facilitate infiltration amplifying the interpreted signal.

54 ENVIRONMENTAL SCIENCES↗

Explainable machine learning for hydrogen diffusion in metals and random binary alloys

Hydrogen diffusion in metals and alloys plays an important role in the discovery of new materials for fuel cell and energy storage technology. While analytic models use hand-selected features that have clear physical ties to hydrogen diffusion, they often lack accuracy when making quantitative predictions. Machine learning models are capable of making accurate predictions, but their inner workings are obscured, rendering it unclear which physical features are truly important. To develop interpretable machine learning models to predict the activation energies of hydrogen diffusion in metals and random binary alloys, we create a database for physical and chemical properties of the species and use it to fit six machine learning models. Our models achieve root-mean-squared errors between 98–119 meV on the testing data and accurately predict that elemental Ru has a large activation energy, while elemental Cr and Fe have small activation energies. By analyzing the feature importances of these fitted models, we identify relevant physical properties for predicting hydrogen diffusivity. While metrics for measuring the individual feature importances for machine learning models exist, correlations between the features lead to disagreement between models and limit the conclusions that can be drawn. Instead grouped feature importance, formed by combining the features via their correlations, agree across the six models and reveal that the two groups containing the packing factor and electronic specific heat are particularly significant for predicting hydrogen diffusion in metals and random binary alloys. In conclusion, this framework allows us to interpret machine learning models and enables rapid screening of new materials with the desired rates of hydrogen diffusion.

36 MATERIALS SCIENCE↗

Towards a framework of nuclear competencies

For the countries considering the introduction of a nuclear energy program, the management of human resources should be a part of the wider integrated management system in order to ensure long term safe and reliable operation. Nuclear energy strategy and approaches to human resources development should take into consideration such fundamental aspects as: development and implementation of a workforce plan, required competencies and qualifications, prerequisites for staffing a nuclear energy program, needed training programs and training facilities, qualification and training requirements. Development of common instruments that respond to the above needs and vision has lead to a new concept of European Credit System for Vocational Education and Training. The European Credit System for Vocational Education and Training (ECVET) is based on definition of 'learning outcomes ' in terms of knowledge, skills and competence, and on identification of portfolios of learning outcomes that allow an individual to prove competencies in a coherent manner. ECVET proposes a common understanding of basic definitions of education and training as well as of the new proposed concepts and it should be recognized by all employers in the EU. In this context, a number of 'Euratom Fission Training Schemes' (EFTS) have been launched in specific areas where a shortage of skilled professionals has been identified. In these schemes the competence building is the result of traditional education plus life-long learning, non-traditional learning, and other forms of educational experiences, relying, in particular, on border-less mobility to get acquainted with various sectors. Each particular Training Scheme should follow a similar path for the achievement of the designed learning outcomes (knowledge, skills, and attitudes). This path to the Training Scheme consists of different activities regarding: definition of training scheme learning outcomes and modules, assessment of prerequisites and student selection, student interview for development of individual training plan, start of the training activities under a specific training scheme. The introduction and recognition of ECVET will lead to a common taxonomy of competencies, and will provide also information about qualifications and units in numerical form, enabling mutual recognition of a training scheme. The description of the learning outcomes to be achieved for qualifying to a specific job profile may follow the analysis phase of the systematic approach to training (SAT). This would ensure a common tool, already used by all trainers. All these steps contribute to establishing of a framework of nuclear competencies recognized and accepted throughout member states. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Metric Learning to Accelerate Convergence of Operator Splitting Methods

Recent developments in machine learning have led to promising advances in accelerating the solution of constrained optimization problems. Increasing demand for real-time decision-making capabilities in applications such as artificial intelligence and optimal control has led to a variety of proposed strategies for learning to produce fast solutions to optimization problems. For example, recent works have shown that it is possible to accelerate the convergence of optimization algorithms by learning to select their parameters, such as gradient descent stepsizes. This work proposes a new approach, in which the underlying metric spaces of proximal operator splitting algorithms are learned to maximize convergence rate. While prior works in optimization theory have derived optimal metrics in simple cases, no such result exists for many practical problem forms including general Quadratic Programming (QP). This paper shows how differentiable optimization can enable the end-to-end learning of proximal metrics, enhancing the convergence of proximal algorithms for QP problems beyond what is possible based on known theory. Additionally, the results illustrate a strong connection between the learned proximal metrics and active constraints at the optima, leading to an interpretation in which the predicted proximal metrics can be viewed as a form of active set prediction.

King, Ethan [BATTELLE (PACIFIC NW LAB)]↗

Explaining Neural Spike Activity for Simulated Bio-plausible Network through Deep Sequence Learning

With significant improvements in large-scale simulations of brain models, there is a growing need to develop tools for rapid analysis and interpreting the simulation results. In this work, we explore the potential of sequential deep learning models to understand and explain the network dynamics among the neurons extracted from a large-scale neural simulation in STACS (Simulation Tool for Asynchronous Cortical Stream). Our method employs a representative neuroscience model that abstracts the cortical dynamics with a reservoir of randomly connected spiking neurons with a low stable spike firing rate throughout the simulation duration. We subsequently analyze the spike dynamics of the simulated spiking neural network through an autoencoder model and an attention-based mechanism.

Kulkarni, Shruti↗

Using High Performance Computing to Examine the Processes of Neurogenesis Underlying Pattern Separation/Completion of Episodic Information.

Using High Performance Computing to Examine the Processes of Neurogenesis Underlying Pattern Separation/Completion of Episodic Information - Sandia researchers developed novel methods and metrics for studying the computational function of neurogenesis, thus generating substantial impact to the neuroscience and neural computing communities. This work could benefit applications in machine learning and other analysis activities.

97 MATHEMATICS AND COMPUTING↗

Enabling the First Ever Measurement of Coherent Neutrino Scattering Through Background Neutron Measurements.

Using High Performance Computing to Examine the Processes of Neurogenesis Underlying Pattern Separation/Completion of Episodic Information - Sandia researchers developed novel methods and metrics for studying the computational function of neurogenesis,thus generating substantial impact to the neuroscience and neural computing communities. This work could benefit applications in machine learning and other analysis activities. The purpose of this project was to computationally model the impact of neural population dynamics within the neurobiological memory system in order to examine how subareas in the brain enable pattern separation and completion of information in memory across time as associated experiences.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Predeployment progress of the Canister Deposition Field Demonstration

This report updates the high-level test plan for evaluating surface deposition on three commercial 32PTH2 spent nuclear fuel (SNF) canisters inside NUTECH Horizontal Modular Storage (NUHOMS) Advanced Horizontal Storage Modules (AHSMs) from Orano (formerly Transnuclear Inc.) and provides a summary of the surface sampling activities that have been conducted to date. The details contained in this report represent the best designs and approaches explored for testing as of this publication. Given the rapidly developing nature of this test program, some of these plans may change to accommodate new objectives or requirements. One goal of this testing is to collect defensible and detailed dust deposition measurements from the surface of dry storage canisters in a marine coastal environment to guide chloride-induced stress corrosion cracking (CISCC) research. Another goal is to provide data for the validation of computational fluid dynamics (CFD) based deposition modeling. To facilitate surface sampling, the otherwise highly prototypic dry storage systems will not contain SNF but rather will be electrically heated to mimic the decay heat and thermal hydraulic environment. Test and heater design is supported by detailed CFD modeling. Instrumentation throughout the canister, storage module, and environment will provide extensive information about the thermal-hydraulic behavior of horizontal dry cask storage systems. Manual sampling over a comprehensive portion of the canister surface at regular time intervals will offer detailed quantification and composition of the deposited particulates from a realistic storage environment. Discussions of a potential host site for the Canister Deposition Field Demonstration (CDFD) are ongoing. Until a host site is chosen, testing of key CDFD hardware components including the heater assemblies, power skid, and remote data acquisition system will continue. Functional testing of the finalized heater assemblies and test apparatus started this fiscal year. These initial heater tests have shown the assemblies are performing within design specifications. Staged surface sampling of a mockup of a canister outside the AHSM on a transfer skid was also performed. Refinements to the sampling procedures and techniques were captured from observation of these activities and lessons-learned debriefs. These updated sampling procedures and techniques are planned to be tested again in the field using the mockup in order to assure personnel are using the most accurate and repeatable methods possible prior to deployment for actual CDFD testing.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Pathway-based analyses of gene expression profiles at low doses of ionizing radiation

Radiation exposure poses a significant threat to human health. Emerging research indicates that even low-dose radiation once believed to be safe, may have harmful effects. This perception has spurred a growing interest in investigating the potential risks associated with low-dose radiation exposure across various scenarios. To comprehensively explore the health consequences of low-dose radiation, our study employs a robust statistical framework that examines whether specific groups of genes, belonging to known pathways, exhibit coordinated expression patterns that align with the radiation levels. Notably, our findings reveal the existence of intricate yet consistent signatures that reflect the molecular response to radiation exposure, distinguishing between low-dose and high-dose radiation. Moreover, we leverage a pathway-constrained variational autoencoder to capture the nonlinear interactions within gene expression data. By comparing these two analytical approaches, our study aims to gain valuable insights into the impact of low-dose radiation on gene expression patterns, identify pathways that are differentially affected, and harness the potential of machine learning to uncover hidden activity within biological networks. This comparative analysis contributes to a deeper understanding of the molecular consequences of low-dose radiation exposure.

63 RADIATION, THERMAL, AND OTHER ENVIRON. POLLUTAN↗

Lessons Learned from Three Mile Island Packaging, Transportation and Disposition that Apply to Fukushima Daiichi Recovery

Following the massive earthquake and resulting tsunami damage in March of 2011 at the Fukushima Daiichi nuclear power plant in Japan, interest was amplified for what was done for recovery at the Three Mile Island Unit 2 (TMI-2) in the United States following its meltdown in 1979. Many parallels could be drawn between to two accidents. This paper presents the results of research done into the TMI-2 recovery effort and its applicability to the Fukushima Daiichi cleanup. This research focused on three topics: packaging, transportation, and disposition. This research work was performed as a collaboration between Japan’s Central Research Institute of Electric Power Industry (CRIEPI) and the Idaho National Laboratory (INL). Hundreds of TMI-2 related documents were searched and pertinent information was gleaned from these documents. Other important information was also obtained by interviewing employees who were involved first hand in various aspects of the TMI-2 cleanup effort. This paper is organized into three main sections: (1) Transport from Three Mile Island to Central Facilities Area at INL, (2) Transport from INL Central Receiving Facility to INL Test Area North (TAN) and wet storage at TAN, and (3) Transport from TAN to INL Idaho Nuclear Technology and Engineering Center (INTEC) and Dry Storage at INTEC. Within each of these sections, lessons learned from performing recovery activities are presented and their applicability to the Fukushima Daiichi nuclear power plant cleanup are outlined.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Fast and Flexible Inference Framework for Continuum Reverberation Mapping Using Simulation-based Inference with Deep Learning

Continuum reverberation mapping (CRM) of active galactic nuclei (AGN) monitors multiwavelength variability signatures to constrain accretion disk structure and supermassive black hole (SMBH) properties. The upcoming Vera Rubin Observatory’s Legacy Survey of Space and Time will survey tens of millions of AGN over the next decade, with thousands of AGN monitored with almost daily cadence in the deep drilling fields. However, existing CRM methodologies often require long computation time and are not designed to handle such large amounts of data. In this paper, we present a fast and flexible inference framework for CRM using simulation-based inference (SBI) with deep learning to estimate SMBH properties from AGN light curves. We use a long short-term memory summary network to reduce the high dimensionality of the light curve data and then use a neural density estimator to estimate the posterior of SMBH parameters. Using simulated light curves, we find SBI can produce more accurate SMBH parameter estimation with 10 3 –10 5 times speed up in inference efficiency compared to traditional methods. The SBI framework is particularly suitable for wide-field CRM surveys as the light curves will have identical observing patterns, which can be incorporated into the SBI simulation. We explore the performance of our SBI model on light curves with irregular-sampled, realistic observing cadence and alternative variability characteristics to demonstrate the flexibility and limitation of the SBI framework.

79 ASTRONOMY AND ASTROPHYSICS↗

High temperature oxidation of corrosion resistant alloys from machine learning

Parabolic rate constants, k p , were collected from published reports and calculated from corrosion product data (sample mass gain or corrosion product thickness) and tabulated for 75 alloys exposed to temperatures between ~800 and 2000 K (~500–1700 °C; 900–3000°F). Data were collected for environments including lab air, ambient and supercritical carbon dioxide, supercritical water, and steam. Materials studied include low- and high-Cr ferritic and austenitic steels, nickel superalloys, and aluminide materials. A combination of Arrhenius analysis, simple linear regression, supervised and unsupervised machine learning methods were used to investigate the relations between composition and oxidation kinetics. The supervised machine learning techniques produced the lowest mean standard errors. The most significant elements controlling oxidation kinetics were Ni, Cr, Al, and Fe, with Mo and Co composition also found to be significant features. The activation energies produced from the machine learning analysis were in the correct distributions for the diffusion constants for the oxide scales expected to dominate in each class.

Materials Science↗