Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “On-the-fly decision making”

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.

13 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Adaptively driven X-ray diffraction guided by machine learning for autonomous phase identification

Machine learning (ML) has become a valuable tool to assist and improve materials characterization, enabling automated interpretation of experimental results with techniques such as X-ray diffraction (XRD) and electron microscopy. Because ML models are fast once trained, there is a key opportunity to bring interpretation in-line with experiments and make on-the-fly decisions to achieve optimal measurement effectiveness, which creates broad opportunities for rapid learning and information extraction from experiments. Here, we demonstrate such a capability with the development of autonomous and adaptive XRD. By coupling an ML algorithm with a physical diffractometer, this method integrates diffraction and analysis such that early experimental information is leveraged to steer measurements toward features that improve the confidence of a model trained to identify crystalline phases. We validate the effectiveness of an adaptive approach by showing that ML-driven XRD can accurately detect trace amounts of materials in multi-phase mixtures with short measurement times. The improved speed of phase detection also enables in situ identification of short-lived intermediate phases formed during solid-state reactions using a standard in-house diffractometer. Our findings showcase the advantages of in-line ML for materials characterization and point to the possibility of more general approaches for adaptive experimentation.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Gaming the beamlines—employing reinforcement learning to maximize scientific outcomes at large-scale user facilities

Abstract Beamline experiments at central facilities are increasingly demanding of remote, high-throughput, and adaptive operation conditions. To accommodate such needs, new approaches must be developed that enable on-the-fly decision making for data intensive challenges. Reinforcement learning (RL) is a domain of AI that holds the potential to enable autonomous operations in a feedback loop between beamline experiments and trained agents. Here, we outline the advanced data acquisition and control software of the Bluesky suite, and demonstrate its functionality with a canonical RL problem: cartpole. We then extend these methods to efficient use of beamline resources by using RL to develop an optimal measurement strategy for samples with different scattering characteristics. The RL agents converge on the empirically optimal policy when under-constrained with time. When resource limited, the agents outperform a naive or sequential measurement strategy, often by a factor of 100%. We interface these methods directly with the data storage and provenance technologies at the National Synchrotron Light Source II, thus demonstrating the potential for RL to increase the scientific output of beamlines, and layout the framework for how to achieve this impact.

36 MATERIALS SCIENCE↗
DOE OSTI2022

An Automated Scanning Transmission Electron Microscope Guided by Sparse Data Analytics

Abstract Artificial intelligence (AI) promises to reshape scientific inquiry and enable breakthrough discoveries in areas such as energy storage, quantum computing, and biomedicine. Scanning transmission electron microscopy (STEM), a cornerstone of the study of chemical and materials systems, stands to benefit greatly from AI-driven automation. However, present barriers to low-level instrument control, as well as generalizable and interpretable feature detection, make truly automated microscopy impractical. Here, we discuss the design of a closed-loop instrument control platform guided by emerging sparse data analytics. We hypothesize that a centralized controller, informed by machine learning combining limited a priori knowledge and task-based discrimination, could drive on-the-fly experimental decision-making. This platform may unlock practical, automated analysis of a variety of material features, enabling new high-throughput and statistical studies.

47 OTHER INSTRUMENTATION↗
DOE OSTI2025

Autonomous alloy composition optimization using molecular dynamics guided by a large language model

Here, we present an autonomous materials discovery framework that couples a large language model (LLM) with molecular dynamics (MD) simulations to optimize Fe–Cr–Mn alloy compositions for tensile strength. Starting from six distinct compositions, the LLM operated as an intelligent agent, iteratively proposing changes based on prior simulation results and constraints. Over 50 iterations per case, the LLM adaptively explored the composition space, identifying high-strength regions, not easily accessible by conventional methods. The highest strength, 18.7 GPa, was achieved with Fe 71 Cr 25 Mn 4 composition, identified from a Fe 75 Cr 20 Mn 5 starting point. The LLM autonomously adjusted its strategy in real time, demonstrating closed-loop decision-making using commodity hardware. This approach showcases the potential of LLMs as scientific co-pilots, capable of accelerating materials discovery and generalizable to other domains like biology and drug design.

Autonomy↗
DOE OSTI2021

Reinforcement Learning Configuration Interaction

Selected configuration interaction (sCI) methods exploit the sparsity of the full configuration interaction (FCI) wave function, yielding significant computational savings and wave function compression without sacrificing the accuracy. Despite recent advances in sCI methods, the selection of important determinants remains an open problem. Furthermore, we explore the possibility of utilizing reinforcement learning approaches to solve the sCI problem. By mapping the configuration interaction problem onto a sequential decision-making process, the agent learns on-the-fly which determinants to include and which to ignore, yielding a compressed wave function at near-FCI accuracy. This method, which we call reinforcement-learned configuration interaction, adds another weapon to the sCI arsenal and highlights how reinforcement learning approaches can potentially help solve challenging problems in electronic structure theory.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRSDate not supplied

Enabling Mission Flexibility to Battery Driven Deep Space Endeavors With Generalized Battery-Health-Monitoring Using Physics-Based and Data-Driven Reduced-Order Models

The needs and requirements for an electrochemical energy storage for deep space exploration is well explored. It is often understood that different mission sites and environmental conditions require different battery chemistries or technologies. Additionally, various engineering solutions are deployed to overcome specific chemical challenges. One often overlooked need is the “health” monitoring of an electrochemical storage system. The term generalized health monitoring, as envisioned in this work, refers to the monitoring of various aspects such as electrode health, electrolyte health, reaction pathway health, cooling system health, sensor health, and BMS health [1]. Generalized health monitoring allows mission leads, engineers, and scientists to incorporate flexibility in mission designs, make on-the-fly mission changes, and extend the duration of science missions. Moreover, it enables automation and data-driven decision-making without compromising safety and performance. Recently, our group developed a hierarchy of thermal reduced-order models (TROM) by combining a physics-based modeling approach and data-driven model reduction techniques applied to flight data [2]. The resulting TROMs were found to be not only accurate but also identifiable from the flight data. Consequently, the coefficient of variance of the model parameters is small over the course of hundreds of flights, allowing for monitoring the parameter evolution trajectories as the battery ages and degrades. These parameters constitute the metrics of the generalized health of a battery. Monitoring their evolution allows such models to be used for anomaly detection and prognostics, improving early detection of abnormal behavior and thus enabling timely maintenance, longer battery life, and enhanced battery safety. For this presentation, the practicality of the thermal model will be validated on a pack of 14cells under various topology configurations such as 1S14P, 2P7S, 7S2P, and 1P14S. It is well known that manufacturing and non-uniform aging lead to variability in the performance of a cell, which is exacerbated by cell balancing during active load. Additionally, in extreme scenarios, the paramount objective is to complete the mission, regardless of the stresses on the battery. Topology-induced balancing issues further stress the battery. The goal of this study is to determine if the noise (identifiability) in the reduced-order thermal model parameters is sensitive to topology, cell spacing, cooling strategy, and manufacturing or age variability. The variability in cells is considered by assuming a multimodal distribution for microscopic parameters of a cell (such as porosity, tortuosity, reaction kinetics, volumetric thermal conductivity, and volumetric heat capacity). The compounded effect of manufacturing variability, topological selection, cooling strategies, and cell balancing ages each cell in a battery differently. The study aims to clarify whether the challenge in extracting maximum information depends on the minimum number of sensors or models used for data extraction.

Automation↗
DOE OSTI2023

Adaptive Ensemble Refinement of Protein Structures in High Resolution Electron Microscopy Density Maps with Radical Augmented Molecular Dynamics Flexible Fitting

Recent advances in cryo-electron microscopy (cryo-EM) have enabled modeling macromolecular complexes that are essential components of the cellular machinery. The density maps derived from cryo-EM experiments are often integrated with manual, knowledge or artificial intelligence driven, and physics-guided computational methods to build, fit, and refine molecular structures. Going beyond a single stationary- structure determination scheme, it is becoming more common to interpret the experimental data with an ensemble of models, which contributes to an average observation. Hence, there is a need to decide on the quality of an ensemble of protein structures on-the-fly, while refining them against the density maps. Here, we introduce such an adaptive decision making scheme during the molecular dynamics flexible fitting (MDFF) of biomolecules. Using RADICAL-Cybertools, and the new RADICAL augmented MDFF implementation (R-MDFF) is examined in high-performance computing environments for refinement of two protein systems, Adenylate Kinase and Carbon Monoxide Dehydrogenase. For the test cases, use of multiple replicas in flexible fitting with adaptive decision making in R-MDFF improves the overall correlation to the density by 40% relative to the refinements of the brute-force MDFF. The improvements are particularly significant at high, 2 - 3 Å, map resolutions. More importantly, the ensemble model captures key features of biologically relevant molecular dynamics that is inaccessible to a single-model interpretation. Finally, the pipeline is applicable to systems of growing sizes, which is demonstrated using ensemble refinement of capsid proteins from Chimpanzee adenovirus. The overhead for decision making remaining low and robust to computing environments. The software is publicly available on GitHub and includes a short user guide to install the R-MDFF on different computing environments, from local Linux based workstations to High Performance Computing (HPC) environments.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRS2004

Autonomous In-Situ Resources Prospector

This presentation will describe the concept of an autonomous, intelligent, rover-based rapid surveying system to identify and map several key lunar resources to optimize their ISRU (In Situ Resource Utilization) extraction potential. Prior to an extraction phase for any target resource, ground-based surveys are needed to provide confirmation of remote observation, to quantify and map their 3-D distribution, and to locate optimal extraction sites (e.g. ore bodies) with precision to maximize their economic benefit. The system will search for and quantify optimal minerals for oxygen production feedstock, water ice, and high glass-content regolith that can be used for building materials. These are targeted because of their utility and because they are, or are likely to be, variable in quantity over spatial scales accessible to a rover (i.e., few km). Oxygen has benefits for life support systems and as an oxidizer for propellants. Water is a key resource for sustainable exploration, with utility for life support, propellants, and other industrial processes. High glass-content regolith has utility as a feedstock for building materials as it readily sinters upon heating into a cohesive matrix more readily than other regolith materials or crystalline basalts. Lunar glasses are also a potential feedstock for oxygen production, as many are rich in iron and titanium oxides that are optimal for oxygen extraction. To accomplish this task, a system of sensors and decision-making algorithms for an autonomous prospecting rover is described. One set of sensors will be located in the wheel tread of the robotic search vehicle providing contact sensor data on regolith composition. Another set of instruments will be housed on the platform of the rover, including VIS-NIR imagers and spectrometers, both for far-field context and near-field characterization of the regolith in the immediate vicinity of the rover. Also included in the sensor suite are a neutron spectrometer, ground-penetrating radar, and an instrumented cone penetrometer for subsurface assessment. Output from these sensors will be evaluated autonomously in real-time by decision-making software to evaluate if any of the targeted resources has been detected, and if so, to quantify their abundance. Algorithms for optimizing the mapping strategy based on target resource abundance and distribution are also included in the autonomous software. This approach emphasizes on-the-fly survey measurements to enable efficient and rapid prospecting of large areas, which will improve the economics of ISRU system approaches. The mature technology will enable autonomous rovers to create in-situ resource maps of lunar or other planetary surfaces, which will facilitate human and robotic exploration.

Dissly, R. W.↗
DOE OSTI2020

CoSim: A Simulator for Co-Scheduling of Batch and On-Demand Jobs in HPC Datacenters

The increasing scale and complexity of scientific applications are rapidly transforming the ecosystem of tools, methods, and workflows adopted by the high-performance computing (HPC) community. Big data analytics and deep learning are gaining traction as essential components in this ecosystem in a variety of scenarios, such as, steering of experimental instruments , acceleration of high-fidelity simulations through surrogate computations, and guided ensemble searches. In this context, the batch job model traditionally adopted by the supercomputing in-frastructures needs to be complemented with support to schedule opportunistic on-demand analytics jobs, leading to the problem of efficient preemption of batch jobs with minimum loss of progress. In this paper, we design and implement a simulator, CoSim, that enables on-the-fly analysis of the trade-offs arising between delaying the start of opportunistic on-demand jobs, which leads to longer analytics latency, and loss of progress due to preemption of batch jobs, which is necessary to make room for on-demand jobs. To this end, we propose an algorithm based on dynamic programming with predictable performance and scalability that enables supercomputing infrastructure schedulers to analyze the aforementioned trade-off and take decisions in near real-time. Compared with other state-of-art approaches using traces of the Theta pre-Exascale machine, our approach is capable of finding the optimal solution, while achieving high performance and scalability.

batch job preemption↗
NASA NTRS2019

Analysis Ready Data in Analytics Optimized Data Stores for Analysis of Big Earth Data in the Cloud

Cloud computing offers the possibility of making the analysis of Big Data approachable for a wider community due to affordable access to computing power, an ecosystem of usable tools for parallel processing, and migration of many large datasets to archives in the cloud, allowing data-proximal computing. Generally, data analysis acceleration in the cloud comes from running multiple nodes in a split-combine-apply strategy. Data systems such as the Earth Observing System Data and Information System are in a position to "pre-split" the data by storing them in a data store that is optimized for data parallel computing, i.e., an Analytics-Optimized Data Store (AODS). A variety of approaches to AODS are possible, from highly scalable databases to scalable filesystems to data formats optimized for cloud access (e.g., zarr and cloud-optimized datasets), with the optimal choice dependent on both the types of analysis and the geospatial structure of the data. A key question is how much preprocessing of the data to do, both before splitting and as the first part of the apply step. Again, the geospatial structure of the data and the analysis type influence the decision, with the added complexity of the user type. Trans-disciplinary users who are not well-versed in the nuances of quality-filtering and georeferencing of remote sensing orbit/swath/scene data tend to ask for more highly processed data, relying on the data provider to make sensible decisions on preprocessing parameters. (This accounts for the popularity of "Level 3" gridded data, despite the lower spatial resolution it provides.) In this case, data can be preprocessed before the split, resulting in higher performance in the rest of the "apply" step, which can be transformative for use cases such as interactive data exploration at scale. Discipline researchers who are experienced with remote sensing data often prefer more flexibility in customizing the preprocessing data into Analysis Ready Data, resulting in more need for on-the-fly preprocessing.

Lynnes, Christopher↗
NASA NTRS2017

GeoDash: Assisting Visual Image Interpretation in Collect Earth Online by Leveraging Big Data on Google Earth Engine

Collect Earth Online (CEO) is a free and open online implementation of the FAO Collect Earth system for collaboratively collecting environmental data through the visual interpretation of Earth observation imagery. The primary collection mechanism in CEO is human interpretation of land surface characteristics in imagery served via Web Map Services (WMS). However, interpreters may not have enough contextual information to classify samples by only viewing the imagery served via WMS, be they high resolution or otherwise. To assist in the interpretation and collection processes in CEO, SERVIR, a joint NASA-USAID initiative that brings Earth observations to improve environmental decision making in developing countries, developed the GeoDash system, an embedded and critical component of CEO. GeoDash leverages Google Earth Engine (GEE) by allowing users to set up custom browser-based widgets that pull from GEE's massive public data catalog. These widgets can be quick looks of other satellite imagery, time series graphs of environmental variables, and statistics panels of the same. Users can customize widgets with any of GEE's image collections, such as the historical Landsat collection with data available since the 1970s, select date ranges, image stretch parameters, graph characteristics, and create custom layouts, all on-the-fly to support plot interpretation in CEO. This presentation focuses on the implementation and potential applications, including the back-end links to GEE and the user interface with custom widget building. GeoDash takes large data volumes and condenses them into meaningful, relevant information for interpreters. While designed initially with national and global forest resource assessments in mind, the system will complement disaster assessments, agriculture management, project monitoring and evaluation, and more.

SERVI↗
DOE OSTI2024

A Deep Learning-Driven Sampling Technique to Explore the Phase Space of an RNA Stem-Loop

The folding and unfolding of RNA stem-loops are critical biological processes; however, their computational studies are often hampered by the ruggedness of their folding landscape, necessitating long simulation times at the atomistic scale. Here, we adapted DeepDriveMD (DDMD), an advanced deep learning-driven sampling technique originally developed for protein folding, to address the challenges of RNA stem-loop folding. Although tempering- and order parameter-based techniques are commonly used for similar rare-event problems, the computational costs or the need for a priori knowledge about the system often present a challenge in their effective use. DDMD overcomes these challenges by adaptively learning from an ensemble of running MD simulations using generic contact maps as the raw input. DeepDriveMD enables on-the-fly learning of a low-dimensional latent representation and guides the simulation toward the undersampled regions while optimizing the resources to explore the relevant parts of the phase space. We showed that DDMD estimates the free energy landscape of the RNA stem-loop reasonably well at room temperature. Our simulation framework runs at a constant temperature without external biasing potential, hence preserving the information on transition rates, with a computational cost much lower than that of the simulations performed with external biasing potentials. Here, we also introduced a reweighting strategy for obtaining unbiased free energy surfaces and presented a qualitative analysis of the latent space. This analysis showed that the latent space captures the relevant slow degrees of freedom for the RNA folding problem of interest. Finally, throughout the manuscript, we outlined how different parameters are selected and optimized to adapt DDMD for this system. We believe this compendium of decision-making processes will help new users adapt this technique for the rare-event sampling problems of their interest.

Gupta, Ayush↗
NASA NTRS2014

Proto-Examples of Data Access and Visualization Components of a Potential Cloud-Based GEOSS-AI System

Once a research or application problem has been identified, one logical next step is to search for available relevant data products. Thus, an early component of a potential GEOSS-AI system, in the continuum between observations and end point research, applications, and decision making, would be one that enables transparent data discovery and access by users. Such a component might be effected via the systems data agents. Presumably, some kind of data cataloging has already been implemented, e.g., in the GEOSS Common Infrastructure (GCI). Both the agents and cataloging could also leverage existing resources external to the system. The system would have some means to accept and integrate user-contributed agents. The need or desirability for some data format internal to the system should be evaluated. Another early component would be one that facilitates browsing visualization of the data, as well as some basic analyses.Three ongoing projects at the NASA Goddard Earth Sciences Data and Information Services Center (GES DISC) provide possible proto-examples of potential data access and visualization components of a cloud-based GEOSS-AI system. 1. Reorganizing data archived as time-step arrays to point-time series (data rods), as well as leveraging the NASA Simple Subset Wizard (SSW), to significantly increase the number of data products available, at multiple NASA data centers, for production as on-the-fly (virtual) data rods. SSWs data discovery is based on OpenSearch. Both pre-generated and virtual data rods are accessible via Web services. 2. Developing Web Feature Services to publish the metadata, and expose the locations, of pre-generated and virtual data rods in the GEOSS Portal and enable direct access of the data via Web services. SSW is also leveraged to increase the availability of both NASA and non-NASA data.3.Federating NASA Giovanni (Geospatial Interactive Online Visualization and Analysis Interface), for multi-sensor data exploration, that would allow each cooperating data center, currently the NASA Distributed Active Archive Centers (DAACs), to configure its own Giovanni deployment, while also allowing all the deployments to incorporate each others data. A federated Giovanni comprises Giovanni Virtual Machines, which can be run on local servers or in the cloud.

access↗