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 1,297 records · Page 72
All sourcesNASA NTRSDOE OSTI
NASA NTRS2003

Reinforcement Learning for Weakly-Coupled MDPs and an Application to Planetary Rover Control

Weakly-coupled Markov decision processes can be decomposed into subprocesses that interact only through a small set of bottleneck states. We study a hierarchical reinforcement learning algorithm designed to take advantage of this particular type of decomposability. To test our algorithm, we use a decision-making problem faced by autonomous planetary rovers. In this problem, a Mars rover must decide which activities to perform and when to traverse between science sites in order to make the best use of its limited resources. In our experiments, the hierarchical algorithm performs better than Q-learning in the early stages of learning, but unlike Q-learning it converges to a suboptimal policy. This suggests that it may be advantageous to use the hierarchical algorithm when training time is limited.

Bernstein, Daniel S.↗
DOE OSTI2023

Mapping Glacier Basal Sliding Applying Machine Learning

During the RESOLVE project (“High-resolution imaging in subsurface geophysics: development of a multi-instrument platform for interdisciplinary research”), continuous surface displacement and seismic array observations were obtained on Glacier d’Argentière in the French Alps for 35 days in May 2018. The data set is used to perform a detailed study of targeted processes within the highly dynamic cryospheric environment. In particular, the physical processes controlling glacial basal motion are poorly understood and remain challenging to observe directly. Especially in the Alpine region for temperate based glaciers where the ice rapidly responds to changing climatic conditions and thus, processes are strongly intermittent in time and heterogeneous in space. Spatially dense seismic and Global Positioning System (GPS) measurements are analyzed applying machine learning to gain insight into the processes controlling glacial motions of Glacier d’Argentière. Using multiple bandpass-filtered copies of the continuous seismic waveforms, we compute energy-based features, develop a matched field beamforming catalog and include meteorological observations. Features describing the data are analyzed with a gradient boosting decision tree model to directly estimate the GPS displacements from the seismic noise. We posit that features of the seismic noise provide direct access to the dominant parameters that drive displacement on the highly variable and unsteady surface of the glacier. The machine learning model infers daily fluctuations and longer term trends. The results show on-ice displacement rates are strongly modulated by activity at the base of the glacier. The techniques presented provide a new approach to study glacial basal sliding and discover its full complexity.

58 GEOSCIENCES↗
NASA NTRS2006

Cassini Distributed Instrument Operations: What We've Learned Since Saturn Orbit Insertion

The Cassini mission to Saturn is complex with 12 science teams conducting distributed operations across the United States and Europe. Each Team includes scientists from around the world who actively participate in operations, including observation design, instrument commanding, downlink processing, and archiving. This represents a change in how JPL complex deep-space missions have been operated. Since Saturn Orbit Insertion (SOI), the Cassini Project has spent 17 months conducting science operations and has gained real-world experience that has tested the assumptions and rationale for this approach. We have learned that many of the expected benefits have been realized, but there were numerous unexpected challenges as well. This paper will discuss the lessons learned from the Cassini Tour experience to date. It will revisit the assumptions and rationale behind the distributed instrument operations design and will describe the results, good and bad, of implementing this method of operations. We will describe how Instrument Teams are structured, their roles and responsibilities, what challenges they faced going into orbital operations (the 'tour') and what creative solutions were proposed when funding limitations and schedule milestones prevented optimum solutions. We will also discuss the problems that have been encountered both on the ground and with the instruments, how these problems and anomalies were overcome, and what was learned along the way about the characteristics of distributed instrument operations.

Cassini↗
DOE OSTI2021

Data-driven Optimal Control Strategy for Virtual Synchronous Generator via Deep Reinforcement Learning Approach

This paper aims at developing a data-driven optimal control strategy for virtual synchronous generator (VSG) in the scenario where no expert knowledge or requirement for system model is available. Firstly, the optimal and adaptive control problem for VSG is transformed into a reinforcement learning task. Specifically, the control variables, i.e., virtual inertia and damping factor, are defined as the actions. Meanwhile, the active power output, angular frequency and its derivative are considered as the observations. Moreover, the reward mechanism is designed based on three preset characteristic functions to quantify the control targets: (1) maintaining the deviation of angular frequency within special limits; (2) preserving well-damped oscillations for both the angular frequency and active power output; (3) obtaining slow frequency drop in the transient process. Next, to maximize the cumulative rewards, a decentralized deep policy gradient algorithm, which features model-free and faster convergence, is developed and employed to find the optimal control policy. With this effort, a data-driven adaptive VSG controller can be obtained. By using the proposed controller, the inverter-based distributed generator can adaptively adjust its control variables based on current observations to fulfill the expected targets in model-free fashion. Finally, simulation results validate the feasibility and effectiveness of the proposed approach.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2025

FracML: A Machine Learning Based Tool to Quantify Reservoir Scale Fracture Network for CO2 Storage

Poster on “FRACML: A Machine Learning Based Tool to Quantify Reservoir Scale Fracture Network for CO2 Storage” for the CCUS 2025 conference held in Houston, Texas March 3-5, 2025. The accurate characterization of subsurface fracture networks is essential for the secure operation of carbon capture, utilization, and storage (CCUS) projects. A thorough understanding of the spatial distribution of subsurface faults and fractures is crucial for predicting CO2 plume evolution and minimizing risks such as potential leakage into overlying formations or induced seismicity. In this context, robust fracture network quantification plays a pivotal role in reservoir management, providing the data necessary to fine-tune operational parameters, and ensure the environmental and economic viability of CCUS projects. As part of the U.S. Department of Energy’s SMART (Science-informed Machine Learning for Accelerating Real-time Decisions in Subsurface Applications) initiative, we focused on the development and application of a machine learning-based tool (FRACML) designed to quantify and map fracture networks using real-world (non-synthetic) data from an active CO2 injection site. Our objective is to demonstrate the utility of this tool in improving operational efficiency and safety across CCUS sites.

artifical intelligence / machine learning (AI/ML)↗
DOE OSTI2023

Experimental Observations of the Topology of Convolutional Neural Network Activations

Topological data analysis (TDA) is a branch of computational mathematics, bridging algebraic topology and data science, that provides compact, noise-robust representations of complex structures. Deep neural networks (DNNs) learn millions of parameters associated with a series of transformations defined by the model architecture resulting in high-dimensional, difficult to interpret internal representations of input data. As DNNs become more ubiquitous across multiple sectors of our society, there is increasing recognition that mathematical methods are needed to aid analysts, researchers, and practitioners in understanding and interpreting how these models' internal representations relate to the final classification. In this paper we apply cutting edge techniques from TDA with the goal of gaining insight towards interpretability of convolutional neural networks used for image classification. We use two common TDA approaches to explore several methods for modeling hidden layer activations as high-dimensional point clouds, and provide experimental evidence that these point clouds capture valuable structural information about the model's process. First, we demonstrate that a distance metric based on persistent homology can be used to quantify meaningful differences between layers and discuss these distances in the broader context of existing representational similarity metrics for neural network interpretability. Second, we show that a mapper graph can provide semantic insight as to how these models organize hierarchical class knowledge at each layer. These observations demonstrate that TDA is a useful tool to help deep learning practitioners unlock the hidden structures of their models.

topological data analysis, deep learning↗
NASA NTRS2008

Automated Data Assimilation and Flight Planning for Multi-Platform Observation Missions

This is a progress report on an effort in which our goal is to demonstrate the effectiveness of automated data mining and planning for the daily management of Earth Science missions. Currently, data mining and machine learning technologies are being used by scientists at research labs for validating Earth science models. However, few if any of these advanced techniques are currently being integrated into daily mission operations. Consequently, there are significant gaps in the knowledge that can be derived from the models and data that are used each day for guiding mission activities. The result can be sub-optimal observation plans, lack of useful data, and wasteful use of resources. Recent advances in data mining, machine learning, and planning make it feasible to migrate these technologies into the daily mission planning cycle. We describe the design of a closed loop system for data acquisition, processing, and flight planning that integrates the results of machine learning into the flight planning process.

Oza, Nikunj↗
DOE OSTI2024

Energy Technology Commercialization and Entrepreneurship: Insights From the U.S. Department of Energy's Office of Technology Transitions 2024 Energy Technology University Prize Faculty Track

The Energy Technology University Prize (EnergyTech UP) was established in 2022 with the goal of challenging student teams to develop impactful business plans for energy technologies of their choosing. EnergyTech UP is part of the The American Made Challenges program portfolio, funded by the U.S. Department of Energy. EnergyTech UP is specifically funded by the Department of Energy's Office of Technology Transfer (OTT) and is administrated by staff at the National Renewable Energy Laboratory (NREL). Since its inception, the annual prize has attracted applications from approximately 1,948 students across nearly every U.S. state and territory, awarding over $1 million in funds to student teams. In 2024, the prize expanded to include a Faculty Track, which invited faculty members from degree-granting institutions across the U.S. to design entrepreneurship-based curricula or educational activities. This new track aims to foster innovation in energy entrepreneurship education by leveraging the expertise of faculty to create robust, practical, and inclusive learning experiences. This report summarizes the themes, strategies and impacts identified in proposals from EnergyTech UP's inaugural Faculty Track. Our primary data for organizing insights are the Faculty Track applications themselves. This evaluation of 2024 EnergyTech UP Faculty Track applications aims to support future EnergyTech UP Faculty Track applicants as well as others who are interested in promoting, developing, and/or implementing educational activities that focus on energy commercialization and entrepreneurship at their institutions. By presenting insights from 2024 entries to the Faculty Track, we hope to contribute to a growing inventory of open-source curriculum development resources and provide materials to facilitate the growth of similar programs at a variety of collegiate institutions.

adoption readiness level↗
DOE OSTI2023

Estimating building occupancy: a machine learning system for day, night, and episodic events

Building occupancy research increasingly emphasizes understanding the social and physical dynamics of how people occupy space. Opportunities in the open source domain including social media, Volunteered Geographic Information, crowdsourcing, and sensor data have proliferated, resulting in the exploration of building occupancy dynamics at varying spatiotemporal scales. At Oak Ridge National Laboratory, research into building occupancies through the development of a global learning framework that accommodates exploitation of open source authoritative sources, including governmental census and surveys, journal articles, real estate databases, and more, to report national and subnational building occupancies across the world continues through the Population Density Tables (PDT) project. This probabilistic learning system accommodates expert knowledge, experience, and open-source data to capture local, socioeconomic, and cultural information about human activity. It does so through a systematic process of data harmonization techniques in the development of observation models for over 50 building types to dynamically update baseline estimates and report probabilistic diurnal and episodic building occupancy estimates. This discussion will explore how PDT is implemented at scale and expanded based on the development of observation model classes and will explain how to interpret and spatially apply the reported probability occupancy estimates and uncertainty.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2023

Dynamic Monkey Bar Mechanism of Superionic Li‐ion Transport in LiTaCl 6

Abstract The LiTaCl 6 solid electrolyte has the lowest activation energy of ionic conduction at ambient conditions (0.165 eV), with a record high ionic conductivity for a ternary compound (11 mS cm −1 ). However, the mechanism has been unclear. We train machine‐learning force fields (MLFF) on ab initio molecular dynamics (AIMD) data on‐the‐fly and perform MLFF MD simulations of AIMD quality up to the nanosecond scale at the experimental temperatures, which allows us to predict accurate activation energy for Li‐ion diffusion (at 0.164 eV). Detailed analyses of trajectories and vibrational density of states show that the large‐amplitude vibrations of Cl − ions in TaCl 6 − enable the fast Li‐ion transport by allowing dynamic breaking and reforming of Li−Cl bonds across the space in between the TaCl 6 − octahedra. We term this process the dynamic‐monkey‐bar mechanism of superionic Li + transport which could aid the development of new solid electrolytes for all‐solid‐state lithium batteries.

Lei, Ming↗
DOE OSTI2023

Dynamic Monkey Bar Mechanism of Superionic Li–ion Transport in LiTaCl 6

Here, the LiTaCl 6 solid electrolyte has the lowest activation energy of ionic conduction at ambient conditions (0.165 eV), with a record high ionic conductivity for a ternary compound (11 mS cm –1 ). However, the mechanism has been unclear. We train machine-learning force fields (MLFF) on ab initio molecular dynamics (AIMD) data on-the-fly and perform MLFF MD simulations of AIMD quality up to the nanosecond scale at the experimental temperatures, which allows us to predict accurate activation energy for Li-ion diffusion (at 0.164 eV). Detailed analyses of trajectories and vibrational density of states show that the large-amplitude vibrations of Cl – ions in TaCl 6 – enable the fast Li-ion transport by allowing dynamic breaking and reforming of Li–Cl bonds across the space in between the TaCl 6 – octahedra. We term this process the dynamic-monkey-bar mechanism of superionic Li + transport which could aid the development of new solid electrolytes for all-solid-state lithium batteries.

25 ENERGY STORAGE↗
DOE OSTI2025

Allosteric prediction via convolutional neural networks and protein structural and dynamical features

Allostery is the phenomenon whereby a binding event or covalent modification at one site in a protein modulates function at a distal site, thus changing a protein’s functional state. As such, it is a ubiquitous aspect of protein functional regulation. Computationally predicting allosteric states is important as part of the broader challenge of functional annotation, but it also has practical implications for drug development, as targeting an allosteric site often affords greater specificity compared with targeting an orthosteric site. This study introduces a machine learning approach to predict the allosteric functional state using the small G-protein KRas as the model system, due to its implication in many types of cancer and being well studied as a result with many x-ray crystallographic structures of KRas available with different mutations and ligands bound. Using structural and dynamical features that can be cast as images, namely interatomic distances, contact maps, covariance, and mutual information, supervised learning was performed using convolutional neural networks. Two pretrained convolutional neural network architectures, GoogLeNet and ResNet18, were fine-tuned to classify KRas into active or inactive states based on these features. Across training regimes, atomic contact maps emerged as the most effective structural feature, whereas linearized mutual information outperformed covariance in capturing dynamical correlations relevant to allostery. Models achieved significant validation accuracy, with atomic contact maps yielding up to 90% accuracy. In conclusion, the findings suggest that integrating global structural rearrangements and correlated motion patterns with deep learning can reliably predict protein allosteric states, offering a promising framework for understanding allosteric regulation and developing targeted therapeutics.

Rajeshwar T., Rajitha [Oak Ridge National Laborato↗
DOE OSTI2025

Predicting receptor-ligand pairing preferences in plant-microbe interfaces via molecular dynamics and machine learning

Microbiome assembly, structure, and dynamics significantly influence plant health. Secreted microbial signaling molecules initiate and mediate symbiosis by binding to structurally compatible plant receptors. For example, lipo-chitooligosaccharides (LCOs), produced by nitrogen-fixing rhizobial bacteria and various fungi, are recognized by plant lysin motif receptor-like kinases (LysM-RLKs), which activate the common symbiotic pathway. Accurately predicting these molecular interactions could reveal complementary signatures underlying the initial stages of endosymbiosis. Despite the breakthrough in protein-ligand structure prediction with deep learning-based tools, such as AlphaFold3, the large size and highly flexible nature of signaling compounds like LCOs present major challenges for detailed structural characterization and binding-affinity prediction. Typical structure-/physics-based methods of ligand virtual screening are designed for small, drug-like molecules, often rely on high-resolution, experimentally determined structures of the protein receptors, and rarely achieve sufficient sampling to obtain converged thermodynamic quantities with large ligands. In this study, we developed a hybrid molecular dynamics/machine learning (MD/ML) approach capable of predicting binding affinity rankings with high accuracy in systems involving large, flexible ligands, despite limited experimental structural information. Using coarse initial structural models, the predictions using the MD/ML workflow achieved strong alignment with experimental trends, particularly in the top-affinity tier for four legume LysM-RLKs (LYR3) binding to LCOs and a chitooligosaccharide. Furthermore, the MD-based conformation selection protocol provided critical structural insights into substrate specificity and binding mechanisms. This study demonstrates a powerful method to screen for challenging cognate ligand-receptors and advance our understanding of the molecular basis of microbial colonization in plants.

Lipo-chitooligosaccharides↗
NASA NTRS1983

A summary of recent developments in transportation hazard classification activities for ammonium perchlorate

The transportation hazard classification of Ammonium Perchlorate is discussed. A test program was completed and data were forwarded to retain a Class 5.1 designation (oxidizer) for AP which is shipped internationally. As a follow-on to the initial team effort to conduct AP tests existing data were examined and a matrix which catalogs test parameters and findings was compiled. A collection of test protocols is developed to standardize test methods for energetic materials of all types. The actions to date are summarized; the participating organizations and their roles as presently understood; specific findings on AP (matrix); and issues, lessons learned, and potential actions of particular interest to the propulsion community which may evolve as a result of future U.N. propellant transportation classification activities.

Koller, A. M., Jr.↗
NASA NTRS2002

Spatial learning and memory is preserved in rats after early development in a microgravity environment

This study evaluated the cognitive mapping abilities of rats that spent part of their early development in a microgravity environment. Litters of male and female Sprague-Dawley rat pups were launched into space aboard the National Aeronautics and Space Administration space shuttle Columbia on postnatal day 8 or 14 and remained in space for 16 days. These animals were designated as FLT groups. Two age-matched control groups remained on Earth: those in standard vivarium housing (VIV) and those in housing identical to that aboard the shuttle (AGC). On return to Earth, animals were tested in three different tasks that measure spatial learning ability, the Morris water maze (MWM), and a modified version of the radial arm maze (RAM). Animals were also tested in an open field apparatus to measure general activity and exploratory activity. Performance and search strategies were evaluated in each of these tasks using an automated tracking system. Despite the dramatic differences in early experience, there were remarkably few differences between the FLT groups and their Earth-bound controls in these tasks. FLT animals learned the MWM and RAM as quickly as did controls. Evaluation of search patterns suggested subtle differences in patterns of exploration and in the strategies used to solve the tasks during the first few days of testing, but these differences normalized rapidly. Together, these data suggest that development in an environment without gravity has minimal long-term impact on spatial learning and memory abilities. Any differences due to development in microgravity are quickly reversed after return to earth normal gravity.

NASA Discipline Neuroscience↗
NASA NTRS2014

EDSN Development Lessons Learned

The Edison Demonstration of Smallsat Networks (EDSN) is a technology demonstration mission that provides a proof of concept for a constellation or swarm of satellites performing coordinated activities. Networked swarms of small spacecraft will open new horizons in astronomy, Earth observations and solar physics. Their range of applications include the formation of synthetic aperture radars for Earth sensing systems, large aperture observatories for next generation telescopes and the collection of spatially distributed measurements of time varying systems, probing the Earths magnetosphere, Earth-Sun interactions and the Earths geopotential. EDSN is a swarm of eight 1.5U Cubesats with crosslink, downlink and science collection capabilities developed by the NASA Ames Research Center under the Small Spacecraft Technology Program (SSTP) within the NASA Space Technology Mission Directorate (STMD). This paper describes the concept of operations of the mission and planned scientific measurements. The development of the 8 satellites for EDSN necessitated the fabrication of prototypes, Flatsats and a total of 16 satellites to support the concurrent engineering and rapid development. This paper has a specific focus on the development, integration and testing of a large number of units including the lessons learned throughout the project development.

Technology Demonstration↗
DOE OSTI2022

Development of a nearly autonomous management and control system for advanced reactors

This presentation discusses the development of a Nearly Autonomous Management and Control (NAMAC) system for advanced reactors. This presentation first introduces the NAMAC-enabled plant control. Next, this presentation introduces major technologies of implementing NAMAC, including operational workflow, digital twin, and advanced machine learning algorithms. To demonstrate the capability of the NAMAC system, NAMAC is connected with a simulator of the Experimental Breeder Reactor II, where NAMAC is activated to provide recommendations during a single loss of flow accident.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
NASA NTRS2021

Scheduling the NASA Deep Space Network with Deep Reinforcement Learning

With three complexes spread evenly across the Earth, NASA’s Deep Space Network (DSN) is the primary means of communications as well as a significant scientific instrument for dozens of active missions around the world. A rapidly rising number of spacecraft and increasingly complex scientific instruments with higher bandwidth requirements have resulted in demand that exceeds the network’s capacity across its 12 antennae. The existing DSN scheduling process operates on a rolling weekly basis and is time-consuming; for a given week, generation of the final baseline schedule of spacecraft tracking passes takes roughly 5 months from the initial requirements submission deadline, with several weeks of peer-to-peer negotiations in between. This paper proposes a deep reinforcement learning (RL) approach to generate candidate DSN schedules from mission requests and spacecraft ephemeris data with demonstrated capability to address real-world operational constraints. A deep RL agent is developed that takes mission requests for a given week as input, and interacts with a DSN scheduling environment to allocate tracks such that its reward signal is maximized. A comparison is made between an agent trained using Proximal Policy Optimization and its random, untrained counterpart. The results represent a proof-of-concept that, given a well-shaped reward signal, a deep RL agent can learn the complex heuristics used by experts to schedule the DSN. A trained agent can potentially be used to generate candidate schedules to bootstrap the scheduling process and thus reduce the turnaround cycle for DSN scheduling.

Wilson, Brian↗