Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Operator 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 757 records · Page 42
All sourcesNASA NTRSDOE OSTI
DOE OSTI2018

Calcined Waste Storage at the Idaho Nuclear Technology and Engineering Center

This comprehensive report provides definitive information regarding the volume, mass, and composition (chemical and radioactivity) of the calcined waste stored at the Idaho Nuclear Technology and Engineering Center on the Idaho National Laboratory Site. The calcine composition data are required for current regulatory compliance and safe operation of the INTEC’s six Calcined Solids Storage Facilities (CSSFs), as well as future retrieval, treatment, and shipping the calcine to an off-Site facility (such as a geologic repository). This report also contains a description of the calcine storage bins that compose each of the CSSFs, emphasizing the features that may be of concern to future calcine retrieval. The CSSFs were designed by different architectural engineering firms and built at different times. Each CSSF has a unique design, reflecting varying design criteria and lessons learned from historical CSSF operation. The varying CSSF designs will affect future calcine retrieval processes and equipment. This revision (5) of the report presents refinements of the databases used to calculate the volume, mass, chemical content, and radioactivity of calcined waste produced and stored within the CSSFs. This revision corrects minor errors and inconsistencies noted in an extensive validation and verification effort of the prior version of the Microsoft Excel workbooks upon which the calcine composition described in this report is based. Historical calcine samples are insufficient in number and scope of analysis to fully characterize the entire inventory of calcine in the CSSFs. Sample data exist for all of the liquid wastes that were calcined. This report provides calcine composition data based on liquid waste sample analyses, the volume of liquid waste calcined, calciner operating data, and CSSF operating data using several large Microsoft Excel (Microsoft 2010) databases and spreadsheets. The Excel spreadsheets are collectively called the Historical Processing Model (HPM). The calcine composition determined by this method compares favorably with historical calcine sample data.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
NASA NTRS2009

Active Learning with Irrelevant Examples

An improved active learning method has been devised for training data classifiers. One example of a data classifier is the algorithm used by the United States Postal Service since the 1960s to recognize scans of handwritten digits for processing zip codes. Active learning algorithms enable rapid training with minimal investment of time on the part of human experts to provide training examples consisting of correctly classified (labeled) input data. They function by identifying which examples would be most profitable for a human expert to label. The goal is to maximize classifier accuracy while minimizing the number of examples the expert must label. Although there are several well-established methods for active learning, they may not operate well when irrelevant examples are present in the data set. That is, they may select an item for labeling that the expert simply cannot assign to any of the valid classes. In the context of classifying handwritten digits, the irrelevant items may include stray marks, smudges, and mis-scans. Querying the expert about these items results in wasted time or erroneous labels, if the expert is forced to assign the item to one of the valid classes. In contrast, the new algorithm provides a specific mechanism for avoiding querying the irrelevant items. This algorithm has two components: an active learner (which could be a conventional active learning algorithm) and a relevance classifier. The combination of these components yields a method, denoted Relevance Bias, that enables the active learner to avoid querying irrelevant data so as to increase its learning rate and efficiency when irrelevant items are present. The algorithm collects irrelevant data in a set of rejected examples, then trains the relevance classifier to distinguish between labeled (relevant) training examples and the rejected ones. The active learner combines its ranking of the items with the probability that they are relevant to yield a final decision about which item to present to the expert for labeling. Experiments on several data sets have demonstrated that the Relevance Bias approach significantly decreases the number of irrelevant items queried and also accelerates learning speed.

Wagstaff, Kiri↗
DOE OSTI2025

Synergistic learning with multi-task DeepONet for efficient PDE problem solving

Multi-task learning (MTL) is an inductive transfer mechanism designed to leverage useful information from multiple tasks to improve generalization performance compared to single-task learning. It has been extensively explored in traditional machine learning to address issues such as data sparsity and overfitting in neural networks. In this work, we apply MTL to problems in science and engineering governed by partial differential equations (PDEs). However, implementing MTL in this context is complex, as it requires task-specific modifications to accommodate various scenarios representing different physical processes. To this end, we present a multi-task deep operator network (MT-DeepONet) to learn solutions across various functional forms of source terms in a PDE and multiple geometries in a single concurrent training session. We introduce modifications in the branch network of the vanilla DeepONet to account for various functional forms of a parameterized coefficient in a PDE. Additionally, we handle parameterized geometries by introducing a binary mask in the branch network and incorporating it into the loss term to improve convergence and generalization to new geometry tasks. Our approach is demonstrated on three benchmark problems: (1) learning different functional forms of the source term in the Fisher equation; (2) learning multiple geometries in a 2D Darcy Flow problem and showcasing better transfer learning capabilities to new geometries; and (3) learning 3D parameterized geometries for a heat transfer problem and demonstrate the ability to predict on new but similar geometries. Finally, our MT-DeepONet framework offers a novel approach to solving PDE problems in engineering and science under a unified umbrella based on synergistic learning that reduces the overall training cost for neural operators.

42 ENGINEERING↗
DOE OSTI2018

SynthNotes: A Generator Framework for High-volume, High-fidelity Synthetic Mental Health Notes

One of the key, emerging challenges that connects the "Big Data" and the AI domain is the availability of sufficient volumes of training data for AI/Machine Learning tasks. SynthNotes is a framework for generating standards-compliant, realistic mental health progress report notes at the very large, population-level scale, and in a strict privacy-preserving manner. Our framework, inspired by the needs to explore, evaluate, and train computational methods for the emerging mental health crisis in the US, is useful for benchmarking, optimization, and training of biomedical natural language processing, information extraction, and machine learning systems intended to operate at "Big Data" scale (billions of notes). The free text notes generated by SynthNotes are based on the literature and public statistical models allowing for realistic, natural language representation of a patient, and his or her mental health characteristics. Additionally, SynthNotes can partially simulate stylistic, grammatical, and expressive characteristics of a licensed mental health professional. SynthNotes is modular and flexible, allowing for representation of variety of conditions, incorporation of alternative foundational models, and parametrization of the variability of the structure, content, and size of the synthetically generated corpus. In this paper, we report on the initial use and performance characteristics of our SynthNotes framework and on the ongoing work for inclusion of content planning and deep learning-based generative methods trained on real data.

Begoli, Edmon↗
DOE OSTI2023

Operator Relaxation and the Optimal Depth of Classical Shadows

Classical shadows are a powerful method for learning many properties of quantum states in a sample-efficient manner, by making use of randomized measurements. Here we study the sample complexity of learning the expectation value of Pauli operators via “shallow shadows,” a recently proposed version of classical shadows in which the randomization step is effected by a local unitary circuit of variable depth t. Here we show that the shadow norm (the quantity controlling the sample complexity) is expressed in terms of properties of the Heisenberg time evolution of operators under the randomizing (“twirling”) circuit—namely the evolution of the weight distribution characterizing the number of sites on which an operator acts nontrivially. For spatially contiguous Pauli operators of weight k, this entails a competition between two processes: operator spreading (whereby the support of an operator grows over time, increasing its weight) and operator relaxation (whereby the bulk of the operator develops an equilibrium density of identity operators, decreasing its weight). From this simple picture we derive (i) an upper bound on the shadow norm which, for depth t~log⁡(k), guarantees an exponential gain in sample complexity over the t=0 protocol in any spatial dimension, and (ii) quantitative results in one dimension within a mean-field approximation, including a universal subleading correction to the optimal depth, found to be in excellent agreement with infinite matrix product state numerical simulations. Our Letter connects fundamental ideas in quantum many-body dynamics to applications in quantum information science, and paves the way to highly optimized protocols for learning different properties of quantum states.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
NASA NTRS1994

Architecture and evolution of Goddard Space Flight Center Distributed Active Archive Center

The Goddard Space Flight Center (GSFC) Distributed Active Archive Center (DAAC) has been developed to enhance Earth Science research by improved access to remote sensor earth science data. Building and operating an archive, even one of a moderate size (a few Terabytes), is a challenging task. One of the critical components of this system is Unitree, the Hierarchical File Storage Management System. Unitree, selected two years ago as the best available solution, requires constant system administrative support. It is not always suitable as an archive and distribution data center, and has moderate performance. The Data Archive and Distribution System (DADS) software developed to monitor, manage, and automate the ingestion, archive, and distribution functions turned out to be more challenging than anticipated. Having the software and tools is not sufficient to succeed. Human interaction within the system must be fully understood to improve efficiency to improve efficiency and ensure that the right tools are developed. One of the lessons learned is that the operability, reliability, and performance aspects should be thoroughly addressed in the initial design. However, the GSFC DAAC has demonstrated that it is capable of distributing over 40 GB per day. A backup system to archive a second copy of all data ingested is under development. This backup system will be used not only for disaster recovery but will also replace the main archive when it is unavailable during maintenance or hardware replacement. The GSFC DAAC has put a strong emphasis on quality at all level of its organization. A Quality team has also been formed to identify quality issues and to propose improvements. The DAAC has conducted numerous tests to benchmark the performance of the system. These tests proved to be extremely useful in identifying bottlenecks and deficiencies in operational procedures.

Bedet, Jean-Jacques↗
NASA NTRS2010

The Value of Identifying and Recovering Lost GN&C Lessons Learned: Aeronautical, Spacecraft, and Launch Vehicle Examples

Within the broad aerospace community the importance of identifying, documenting and widely sharing lessons learned during system development, flight test, operational or research programs/projects is broadly acknowledged. Documenting and sharing lessons learned helps managers and engineers to minimize project risk and improve performance of their systems. Often significant lessons learned on a project fail to get captured even though they are well known 'tribal knowledge' amongst the project team members. The physical act of actually writing down and documenting these lessons learned for the next generation of NASA GN&C engineers fails to happen on some projects for various reasons. In this paper we will first review the importance of capturing lessons learned and then will discuss reasons why some lessons are not documented. A simple proven approach called 'Pause and Learn' will be highlighted as a proven low-impact method of organizational learning that could foster the timely capture of critical lessons learned. Lastly some examples of 'lost' GN&C lessons learned from the aeronautics, spacecraft and launch vehicle domains are briefly highlighted. In the context of this paper 'lost' refers to lessons that have not achieved broad visibility within the NASA-wide GN&C CoP because they are either undocumented, masked or poorly documented in the NASA Lessons Learned Information System (LLIS).

Dennehy, Cornelius J.↗
DOE OSTI2022

Learning Optimal Multigrid Smoothers via Neural Networks

Multigrid methods are one of the most efficient techniques for solving large sparse linear systems arising from partial differential equations (PDEs) and graph Laplacians from machine learning applications. One of the key components of multigrid is smoothing, which aims at reducing high-frequency errors on each grid level. However, finding optimal smoothing algorithms is problem-dependent and can impose challenges for many problems. In this paper, we propose an efficient adaptive framework for learning optimized smoothers from operator stencils in the form of convolutional neural networks (CNNs). Here, the CNNs are trained on small-scale problems from a given type of PDEs based on a supervised loss function derived from multigrid convergence theories and can be applied to large-scale problems of the same class of PDEs. Numerical results on anisotropic rotated Laplacian problems and variable coefficient diffusion problems demonstrate improved convergence rates and solution time compared with classical hand-crafted relaxation methods.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

"Reinforcement Learning Based Oscillation Dampening: Scaling up Single-Agent RL algorithms to a 100 AV highway field operational test"

In this article, we explore the technical details of the reinforcement learning (RL) algorithms that were deployed in the largest field test of automated vehicles designed to smooth traffic flow in history as of 2023, uncovering the challenges and breakthroughs that come with developing RL controllers for automated vehicles. We delve into the fundamental concepts behind RL algorithms and their application in the context of self-driving cars, discussing the developmental process from simulation to deployment in detail, from designing simulators to reward function shaping. We present the results in both simulation and deployment, discussing the flow-smoothing benefits of the RL controller. From understanding the basics of Markov decision processes to exploring advanced techniques such as deep RL, our article offers a comprehensive overview and deep dive of the theoretical foundations and practical implementations driving this rapidly evolving field. We also showcase real-world case studies and alternative research projects that highlight the impact of RL controllers in revolutionizing autonomous driving. From tackling complex urban environments to dealing with unpredictable traffic scenarios, these intelligent controllers are pushing the boundaries of what automated vehicles can achieve. Furthermore, we examine the safety considerations and hardware-focused technical details surrounding deployment of RL controllers into automated vehicles. As these algorithms learn and evolve through interactions with the environment, ensuring their behavior aligns with safety standards becomes crucial. We explore the methodologies and frameworks being developed to address these challenges, emphasizing the importance of building reliable control systems for automated vehicles.

Jang, Kathy↗
NASA NTRS1992

The application of artificial intelligence to astronomical scheduling problems

Efficient utilization of expensive space- and ground-based observatories is an important goal for the astronomical community; the cost of modern observing facilities is enormous, and the available observing time is much less than the demand from astronomers around the world. The complexity and variety of scheduling constraints and goals has led several groups to investigate how artificial intelligence (AI) techniques might help solve these kinds of problems. The earliest and most successful of these projects was started at Space Telescope Science Institute in 1987 and has led to the development of the Spike scheduling system to support the scheduling of Hubble Space Telescope (HST). The aim of Spike at STScI is to allocate observations to timescales of days to a week observing all scheduling constraints and maximizing preferences that help ensure that observations are made at optimal times. Spike has been in use operationally for HST since shortly after the observatory was launched in Apr. 1990. Although developed specifically for HST scheduling, Spike was carefully designed to provide a general framework for similar (activity-based) scheduling problems. In particular, the tasks to be scheduled are defined in the system in general terms, and no assumptions about the scheduling timescale are built in. The mechanisms for describing, combining, and propagating temporal and other constraints and preferences are quite general. The success of this approach has been demonstrated by the application of Spike to the scheduling of other satellite observatories: changes to the system are required only in the specific constraints that apply, and not in the framework itself. In particular, the Spike framework is sufficiently flexible to handle both long-term and short-term scheduling, on timescales of years down to minutes or less. This talk will discuss recent progress made in scheduling search techniques, the lessons learned from early HST operations, the application of Spike to other problem domains, and plans for the future evolution of the system.

Johnston, Mark D.↗
DOE OSTI2024

Improving the Freight Productivity of a Heavy-Duty, Battery Electric Truck by Intelligent Energy Management

This project aimed to enhance the range and reduce the operating costs of battery electric Class 8 trucks traveling over 250 miles daily. This was achieved through the development and implementation of an intelligent-Energy Management System (i-EMS) that leverages vehicle and operations data, physics-aware machine learning algorithms, and vehicle-to-cloud (V2C) connectivity. The project hypothesized that advanced machine learning algorithms and real-time data analytics could significantly improve the energy efficiency and range of these trucks. Key objectives included developing a physics-aware machine learning algorithm, implementing an i-EMS with V2C connectivity and physics-aware spatial data analytics (PSDA), and validating the system’s effectiveness with fleet partners HEB Companies and Murphy Logistics. Extensive data collection from vehicle operations, including vehicle characteristics, road conditions, and payload, was conducted. A machine learning algorithm was developed to predict energy consumption and enable proactive decision-making. The i-EMS was implemented on two Volvo VNR BEVs, with operators receiving charging and routing recommendations. Charging stations were installed at depot locations in Texas and Minnesota, with an additional on-route charger in Minnesota. Significant findings included a 14% range improvement for Murphy Logistics on a highway-driving eco-route and a 22% range improvement for HEB Companies on a city-driving eco-route. The i-EMS utilized rule-based methods and physics-based algorithms to predict and reduce energy consumption, with real-time monitoring and analysis through V2C connectivity enabling proactive decision-making. The project demonstrated the feasibility and economic viability of battery electric Class 8 trucks for long-haul operations, showcasing the potential of physics-aware machine learning in optimizing energy management. The successful implementation of the i-EMS in real-world scenarios validates its practical application and effectiveness, paving the way for the widespread adoption of battery electric vehicles in the freight transportation industry.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2025

Machine learning for reactor power monitoring with limited labeled data

Real-time reactor power monitoring is critical for a variety of nuclear applications, spanning safety, security, operations, and maintenance. While machine learning methods have shown promise in monitoring reactor power levels, there is limited research on their efficacy in label-starved environments. The goal of this work is to assess the feasibility of classifying nuclear reactor power level using multisource data in scenarios with limited labels. Data were collected using low-resolution multisensors at four nuclear reactor facilities: two large research reactors and two TRIGA reactors. Within each pair, one reactor dataset served as the source and the other as the target in a transfer learning paradigm. Twenty-three supervised models were trained on labeled sequences of magnetic field and acceleration data from each of the target sites. Self-learning and transfer learning methods were applied to the top performing models to assess their classification performance with increasing amounts of labeled data. While reactor power level classification was achieved with a Matthews Correlation Coefficient of up to 0.739 ± 0.003 and 0.622 ± 0.009 with only 400 sequences per power state for the large research reactor and TRIGA target sites, respectively, self-learning and transfer learning leveraging source site data did not improve target classification performance. These findings suggest that alternative methods, such as higher sensitivity sensors, digital twins, or the use of physics-informed models, are required to enable high-performance classification in machine learning approaches to reactor monitoring with a dearth of target ground truth.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
NASA NTRS2003

Operational thermal control of Cassini Titan flybys

This paper focuses on the technical thermal control evaluation and strategy, the systems-level approach taken, and lessons learned and recommendations in an operations environment.

Cassini Titan flyby thermal analysis↗
DOE OSTI2022

Atikokan Digital Twin: Machine learning in a biomass energy system

The Atikokan Generating Station, operated by Ontario Power Generation, has a 200 MW, biomass-fired tower boiler that operates on a dispatch schedule with a five-minute cycle. The boiler is generally operated in the range of 40–100 MW using two of five burner levels. In order to optimize boiler performance, we propose the implementation of a unique digital twin. Our digital twin abstraction couples Bayesian inference from science-based models and from observations (machine learning) with decision theory to predict operating-variable set points that optimize the physical asset (the boiler) in the presence of uncertainty (artificial intelligence). We focus this paper on the continuous Bayesian machine learning part of the Atikokan Digital Twin; we discuss decision theory in a companion paper. We identify and learn about 12 operational, model, and measured-output parameters and their uncertainties from high-fidelity, science-based simulations of the Atikokan boiler and from the observed measurements at the power plant. Since the goal of the Atikokan Digital Twin is to implement it online in real time, we require fast function evaluations for the quantities of interest extracted from the simulations in the Bayesian analysis. We use Gaussian process regression/interpolation to create accurate, robust surrogate models. We define the Bayesian priors and likelihood function and solve for the posterior distributions of the 12 parameters. Here we then propagate these distributions (i.e., parameters with uncertainty) into the predicted distributions of 790 quantities of interest to learn about the relative importance of various sources of error including experimental, model, and operating-parameter errors.

09 BIOMASS FUELS↗
NASA NTRS2011

Software Architecture to Support the Evolution of the ISRU RESOLVE Engineering Breadboard Unit 2 (EBU2)

The In-Situ Resource Utilization (ISRU) Regolith & Environmental Science and Oxygen & Lunar Volatiles Extraction (RESOLVE) software provides operation of the physical plant from a remote location with a high-level interface that can access and control the data from external software applications of other subsystems. This software allows autonomous control over the entire system with manual computer control of individual system/process components. It gives non-programmer operators the capability to easily modify the high-level autonomous sequencing while the software is in operation, as well as the ability to modify the low-level, file-based sequences prior to the system operation. Local automated control in a distributed system is also enabled where component control is maintained during the loss of network connectivity with the remote workstation. This innovation also minimizes network traffic. The software architecture commands and controls the latest generation of RESOLVE processes used to obtain, process, and quantify lunar regolith. The system is grouped into six sub-processes: Drill, Crush, Reactor, Lunar Water Resource Demonstration (LWRD), Regolith Volatiles Characterization (RVC) (see example), and Regolith Oxygen Extraction (ROE). Some processes are independent, some are dependent on other processes, and some are independent but run concurrently with other processes. The first goal is to analyze the volatiles emanating from lunar regolith, such as water, carbon monoxide, carbon dioxide, ammonia, hydrogen, and others. This is done by heating the soil and analyzing and capturing the volatilized product. The second goal is to produce water by reducing the soil at high temperatures with hydrogen. This is done by raising the reactor temperature in the range of 800 to 900 C, causing the reaction to progress by adding hydrogen, and then capturing the water product in a desiccant bed. The software needs to run the entire unit and all sub-processes; however, throughout testing, many variables and parameters need to be changed as more is learned about the system operation. The Master Events Controller (MEC) is run on a standard laptop PC using Windows XP. This PC runs in parallel to another laptop that monitors the GC, and a third PC that monitors the drilling/ crushing operation. These three PCs interface to the process through a CompactRIO, OPC Servers, and modems.

Moss, Thomas↗
NASA NTRS2011

Generalized Software Architecture Applied to the Continuous Lunar Water Separation Process and the Lunar Greenhouse Amplifier

This innovation provides the user with autonomous on-screen monitoring, embedded computations, and tabulated output for two new processes. The software was originally written for the Continuous Lunar Water Separation Process (CLWSP), but was found to be general enough to be applicable to the Lunar Greenhouse Amplifier (LGA) as well, with minor alterations. The resultant program should have general applicability to many laboratory processes (see figure). The objective for these programs was to create a software application that would provide both autonomous monitoring and data storage, along with manual manipulation. The software also allows operators the ability to input experimental changes and comments in real time without modifying the code itself. Common process elements, such as thermocouples, pressure transducers, and relative humidity sensors, are easily incorporated into the program in various configurations, along with specialized devices such as photodiode sensors. The goal of the CLWSP research project is to design, build, and test a new method to continuously separate, capture, and quantify water from a gas stream. The application is any In-Situ Resource Utilization (ISRU) process that desires to extract or produce water from lunar or planetary regolith. The present work is aimed at circumventing current problems and ultimately producing a system capable of continuous operation at moderate temperatures that can be scaled over a large capacity range depending on the ISRU process. The goal of the LGA research project is to design, build, and test a new type of greenhouse that could be used on the moon or Mars. The LGA uses super greenhouse gases (SGGs) to absorb long-wavelength radiation, thus creating a highly efficient greenhouse at a future lunar or Mars outpost. Silica-based glass, although highly efficient at trapping heat, is heavy, fragile, and not suitable for space greenhouse applications. Plastics are much lighter and resilient, but are not efficient for absorbing longwavelength infrared radiation and therefore will lose more heat to the environment compared to glass. The LGA unit uses a transparent polymer antechamber that surrounds part of the greenhouse and encases the SGGs, thereby minimizing infrared losses through the plastic windows. With ambient temperatures at the lunar poles at 50 C, the LGA should provide a substantial enhancement to currently conceived lunar greenhouses. Positive results obtained from this project could lead to a future large-scale system capable of running autonomously on the Moon, Mars, and beyond. The software for both applications needs to run the entire units and all subprocesses; however, throughout testing, many variables and parameters need to be changed as more is learned about the system operation. The software provides the versatility to permit the software operation to change as the user requirements evolve.

Perusich, Stephen↗
DOE OSTI2025

Self-adaptive weights based on balanced residual decay rate for physics-informed neural networks and deep operator networks

Physics-informed deep learning has emerged as a promising alternative for solving partial differential equations. However, for complex problems, training these networks can still be challenging, often resulting in unsatisfactory accuracy and efficiency. In this work, we demonstrate that the failure of plain physics-informed neural networks arises from the significant discrepancy in the convergence rate of residuals at different training points, where the slowest convergence rate dominates the overall solution convergence. Based on these observations, we propose a pointwise adaptive weighting method that balances the residual decay rate across different training points. The performance of our proposed adaptive weighting method is compared with current state-of-the-art adaptive weighting methods on benchmark problems for both physics-informed neural networks and physics-informed deep operator networks. In conclusion, through extensive numerical results we demonstrate that our proposed approach of balanced residual decay rates offers several advantages, including bounded weights, high prediction accuracy, fast convergence rate, low training uncertainty, low computational cost, and ease of hyperparameter tuning.

Balanced convergence rate↗