Search NASA⌕ Search

SEARCH · Search NASA

Results for “machine learning operations”

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 829 records · Page 46

The Anatomy of Software Changes and Bugs in Autonomous Operating System

Cyberphysical systems with autonomous functions are complex pieces of software, consisting of many components, some of which implement autonomous functionality and some may use AI or machine learning algorithms. Software bugs in an autonomous system are of particular concern, as they can have catastrophic consequences. However, detailed studies based on empirical data are rare and therefore these bugs are not well understood. This paper aims to contribute towards filling that gap by investigating the software changes and bugs in Autonomy Operating System (AOS) for Unmanned Aircraft Systems (UAS), which consist of 26 components containing about 103,000 lines of code and having a total of 772 bugfixes. Based on the data extracted from the code repository and semi-structured interviews with the developers of AOS, we explore the differences among autonomous software components, components developed using Model-based Software Engineering, and reuse with respect to change proneness, fault proneness, distribution of bugfixes among AOS components and files of these components, and characteristics of bugs of different AOS components. Our results show that the autonomous components were significantly more change prone (measured in number of commits and code churn) and fault prone (measured in bugfixes per KLoC) than non-autonomous components. The distribution of the locations of bugfixes was skewed, both at component and file level (i.e., a small number of components / files contained the majority of bugs). These evidence-based findings provide important insights to researchers and practitioners alike and can be used to efficiently improve the quality and reliability of autonomous systems.

Katerina Goseva-Popstojanova↗

Outage Forecast-Based Preventative Scheduling Model for Distribution System Resilience Enhancement: Preprint

Distribution system resilience enhancement is an important topic to ensure customers have access to the power supply during extreme events. In fact, certain weather-related extreme events can be predicted ahead of time. Therefore, it is important to investigate how to predict grid outages using extreme weather forecasts, and how outage predictions can be incorporated into distribution system resilience enhancement. In this paper, a preventative scheduling model for distribution systems is proposed. The model targets at allocating resources, especially mobile responsive resources such as mobile backup generators and mobile energy storage systems, to prepare for an extreme event in the day-ahead context. To achieve efficient resource allocation and scheduling, a machine learning-based outage prediction module is developed to predict vulnerable or risky segments of the distribution system based on historical operating records and extreme weather event forecasts. By integrating the outage prediction results into the scheduling model, optimal resource allocation can be derived to help distribution systems prepare for an upcoming event and improve resilience performance. A real distribution feeder in North Carolina, U.S. is used in the case study to validate the proposed approach.

distributed energy resources↗

Outage Forecast-Based Preventative Scheduling Model for Distribution System Resilience Enhancement

Distribution system resilience enhancement is an important topic to ensure customers have access to power supply during extreme events. In fact, certain weather-related extreme events can be predicted ahead of time. Therefore, it is important to investigate how to predict grid outages using extreme weather forecasts, and how outage predictions can be incorporated into distribution system resilience enhancement. In this paper, a preventative scheduling model for distribution systems is proposed. The model targets at allocating resources, especially mobile responsive resources such as mobile backup generators and mobile energy storage systems, to prepare for an extreme event in the day-ahead context. To achieve efficient resource allocation and scheduling, a machine learning-based outage prediction module is developed to predict vulnerable or risky segments of the distribution system based on historical operating records and extreme weather event forecast. By integrating the outage prediction results into the scheduling model, optimal resource allocation can be derived to help distribution systems prepare for an upcoming event and improve resilience performance. A real distribution feeder in North Carolina, U.S. is used in the case study to validate the proposed approach.

distributed energy resources↗

Deployment of Dynamic Neural Network Optimization to Minimize Heat Rate During Ramping for Coal Power Plants (Final Technical Report)

Much success was achieved throughout the course of this project. A successful implementation of Dynamic Neural Network Optimization (D-NNO) was coupled with Adaptive Predictive Controls (APC) and a novel hardware installation comprised of an advanced sensor network (ASN) measuring mass-weighted averages of flue gas constituents above the horizontal superheater of a coal-fired utility boiler. From 2019 through 2023 (including an extension due to COVID delays), the team was able to prototype, evaluate, deploy, iterate, and ultimately finalize an advanced closed-loop control D-NNO system which demonstrated the ability to: •improve unit efficiency ~2.0% relative to unoptimized operation (represented as total fuel fired per MWh generated) •improve unit NOx emission rates 10%+ beyond static optimization baselines •improve unit temperature stability as much as 58% and on average 12% •improve operating load stability as much as 35% The culmination of this project has generated an advanced methodology of deploying specially designed recurrent neural networks (long short-term memory, gated recurrent unit, encoder-decoder networks, transformers, etc.), customized trajectory planning and closed-loop optimization modules capable of adapting to live electric grid responses and demands, self-tuning and adaptive expert controls constantly adjusting prediction parameters to real-time unit behavior, and a hardware/software package able to reliably calculate net unit heat rate (NUHR) in real-time using flue gas constituents, machine learning, and known combustion relationships. Through this real-time NUHR value, immediate feedback on system adjustments relative to operating efficiency was available, allowing for rapid improvements to system performance. In addition to development and deployment of the advanced D-NNO system, the approach methodology has been readily commercialized through the project platform Griffin Open Systems, LLC, the D-NNO software platform host. Similar methodologies to those developed by this project have already been deployed at 5 other units across the United States, with another 6 implementations scheduled, and more expected. Over the course of the project, multiple academic papers were submitted and accepted for publication within esteemed academic journals, and PhD students were trained and graduated, as well as undergraduate students becoming involved and participating to project objectives.

01 COAL, LIGNITE, AND PEAT↗

Machine learning molecular dynamics simulations toward exploration of high-temperature properties of nuclear fuel materials: case study of thorium dioxide

Predicting materials properties of nuclear fuel compounds is a challenging task in materials science. Their thermodynamical behaviors around and above the operational temperature are essential for the design of nuclear reactors. However, they are not easy to measure, because the target temperature range is too high to perform various standard experiments safely and accurately. Moreover, theoretical methods such as first-principles calculations also suffer from the computational limitations in calculating thermodynamical properties due to their high calculation-costs and complicated electronic structures stemming from f-orbital occupations of valence electrons in actinide elements. Here, we demonstrate, for the first time, machine-learning molecular-dynamics to theoretically explore high-temperature thermodynamical properties of a nuclear fuel material, thorium dioxide. The target compound satisfies first-principles calculation accuracy because f-electron occupation coincidentally diminishes and the scheme meets sampling sufficiency because it works at the computational cost of classical molecular-dynamics levels. We prepare a set of training data using first-principles molecular dynamics with small number of atoms, which cannot directly evaluate thermodynamical properties but captures essential atomistic dynamics at the high temperature range. Then, we construct a machine-learning molecular-dynamics potential and carry out large-scale molecular-dynamics calculations. Consequently, we successfully access two kinds of thermodynamic phase transitions, namely the melting and the anomalous λ transition induced by large diffusions of oxygen atoms. Furthermore, we quantitatively reproduce various experimental data in the best agreement manner by selecting a density functional scheme known as SCAN. Our results suggest that the present scale-up simulation-scheme using machine-learning techniques opens up a new pathway on theoretical studies of not only nuclear fuel compounds, but also a variety of similar materials that contain both heavy and light elements, like thorium dioxide.

74 ATOMIC AND MOLECULAR PHYSICS↗

Matrix Multiply Performance of GPUs on Exascale-class HPE/Cray Systems

The computation of dense matrix-matrix products (GEMMs) is central to many modeling and simulation workloads as well as AI/ML deep learning campaigns. In fact, millions of dollars are spent annually on computing GEMMs, and large model training demands are increasing exponentially. Specialized processors such as GPUs are designed to perform well for these operations. However, the performance of GEMMs on GPUs can exhibit complex behaviors depending on many factors, making it challenging to optimize the performance of GEMMs on these processors. In this study we undertake an examination of GEMM performance on several leading GPU models taken from product lines of GPUs to be deployed in forthcoming exascale computing systems. We show results to illustrate the many factors that can affect performance of GEMMs on GPUs. We then present data collected from a large number of test runs for an example GEMM operation to show the dependence behaviors of GEMM rate on matrix dimensions. Finally, we show results from machine learning-based performance models using novel feature engineering methods to fit the measured performance, providing a potential basis for GEMM performance tuning and autotuning methods for GPUs. Recommendations are also given for how to achieve high GEMM performance on modern GPUs.

Melesse Vergara, Veronica↗

Geophysical Observations Toolkit For Evaluating Coral Health (GOTECH) Fall 2021 Final Report

The NASA Langley Research Center (LaRC) Data Science Team (DST), under the Office of the Chief Information Officer (OCIO), is investigating the capacity of the Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) satellite to infer the vitality of coral reefs. This report describes the Fall 2021 period of performance for the Geophysical Observations Toolkit for Evaluating Coral Health (GOTECH) project. During this effort, two student teams at Georgia Tech developed machine-learning models to predict the vitality of coral reefs in targeted geographic regions based on backscatter data from the CALIPSO satellite. To train these models, students fused data to form a common operating picture of how coral reefs have grown and decayed worldwide. This report describes the student assignment, background, and results of the semester's research.

Machine Learning↗

Fourier-based three-dimensional multistage transformer for aberration correction in multicellular specimens

High-resolution tissue imaging is often compromised by sample-induced optical aberrations that degrade resolution and contrast. Although wavefront sensor-based adaptive optics (AO) can measure these aberrations, such hardware solutions are typically complex, expensive to implement and slow when serially mapping spatially varying aberrations across large fields of view. Here we introduce AOViFT (adaptive optical vision Fourier transformer)—a machine learning-based aberration sensing framework built around a three-dimensional multistage vision transformer that operates on Fourier domain embeddings. AOViFT infers aberrations and restores diffraction-limited performance in puncta-labeled specimens with substantially reduced computational cost, training time and memory footprint compared to conventional architectures or real-space networks. We validated AOViFT on live gene-edited zebrafish embryos, demonstrating its ability to correct spatially varying aberrations using either a deformable mirror or postacquisition deconvolution. By eliminating the need for the guide star and wavefront sensing hardware and simplifying the experimental workflow, AOViFT lowers technical barriers for high-resolution volumetric microscopy across diverse biological samples.

Alshaabi, Thayer [Howard Hughes Medical Institute,↗

Device Feasibility Analysis of Multi-level FeFETs for Neuromorphic Computing

As an emerging non-volatile memory device technology, Ferroelectric Field-Effect Transistors (FeFETs) can enable low-power, adaptive intelligent system design. However, device dimension and operating voltage dependent reliability issues of scaled FeFETs can ultimately lead to degraded performance in solving machine learning tasks. In this article, detailed experimental characterization of FeFET devices of different dimensions have been carried out to explicitly evaluate the non-ideal behavior in device conductance programming properties like number of programming states, cycle-to-cycle (C2C) variations, device-to-device (D2D) variations, and state retention. A hardware-aware software simulation approach has been adopted to capture the adversarial effects of the non-idealities on recognition accuracy through algorithm-level performance assessment by including them in NeuroSim, a popular neural network hardware simulator, to execute a neural network model considering all other hardware constraints. With the added non-idealities, significant accuracy degradation has been observed compared to the ideal scenarios where D2D variations play the most critical role. Thereafter, feasibility of a variation-aware training method has been evaluated to tackle the accuracy drop.

42 ENGINEERING↗

Using MLIR Framework for Codesign of ML Architectures Algorithms and Simulation Tools

MLIR (Multi-Level Intermediate Representation), is an extensible compiler framework that supports high-level data structures and operation constructs. These higher-level code representations are particularly applicable to the artificial intelligence and machine learning (AI/ML) domain, allowing developers to more easily support upcoming heterogeneous AI/ML accelerators and develop flexible domain specific compilers/frameworks with higher-level intermediate representations (IRs) and advanced compiler optimizations. The result of using MLIR within the LLVM compiler framework is expected to yield significant improvement in the quality of generated machine code, which in turn will result in improved performance and hardware efficiency

97 MATHEMATICS AND COMPUTING↗

High Value Opportunities to Advance Automation in Electric Grid Control Rooms

Grid operators and electric utilities are increasingly driven to develop automated processes and introduce decision support tools to reduce cognitive demand on System Operations staff and aid in the operational decision-making process. However, System Operation staff are traditionally hesitant to implement and accept new solutions, automation processes, and tools within the control room. This paper explores high-value opportunities to advance automation in electric grid control rooms for the purpose of improving grid safety, reliability, and resiliency during normal and emergency operating conditions.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Smart Congestion Control for Delay- and Disruption Tolerant Networks

In this paper, we propose a novel congestion control framework for delay- and disruption tolerant networks (DTNs). The proposed framework, called Smart-DTN-CC, adjusts its operation automatically as a function of the dynamics of the underlying network. It employs reinforcement learning, a machine learning technique known to be well suited to problems in which the environment, in this case the network, plays a crucial role; yet, no prior knowledge about the target environment can be assumed, i.e., the only way to acquire information about the environment is to interact with it through continuous online learning. Smart-DTN-CC nodes get input from the environment (e.g., its buffer occupancy, set of neighbors, etc), and, based on that information, choose an action to take from a set of possible actions. Depending on an action’s effectiveness in controlling congestion, it will be given a reward. Smart-DTN-CC’s goal is to maximize the overall reward which translates to minimizing congestion. To our knowledge, Smart-DTN-CC is the first DTN congestion control framework that has the ability to automatically and continuously adapt to the dynamics of the target environment which allows Smart-DTNCC to deliver adequate performance in a variety of DTN applications and scenarios. As demonstrated by our experimental evaluation, Smart-DTN-CC is able to consistently outperform existing DTN congestion control mechanisms under a wide range of network conditions and characteristics.

Hirata, Celso M.↗

Neural networks for modeling and control of particle accelerators

Charged particle accelerators support a wide variety of scientific, industrial, and medical applications. They range in scale and complexity from systems with just a few components for beam acceleration and manipulation, to large scientific user facilities that span many kilometers and have hundreds-to-thousands of individually-controllable components. Specific operational requirements must be met by adjusting the many controllable variables of the accelerator. Meeting these requirements can be challenging, both in terms of the ability to achieve specific beam quality metrics in a reliable fashion and in terms of the time needed to set up and maintain the optimal operating conditions. One avenue toward addressing this challenge is to incorporate techniques from the fields of machine learning (ML) and artificial intelligence (AI) into the way particle accelerators are modeled and controlled. While many promising approaches within AI/ML could be used for particle accelerators, this dissertation focuses on approaches based on neural networks. Neural networks are particularly well-suited to modeling, control, and diagnostic analysis of nonlinear systems, as well as systems with large parameter spaces. They are also very appealing for their ability to process high-dimensional data types, such as images and time series (both of which are ubiquitous in particle accelerators). In this work, key studies that demonstrated the potential utility of modern neural network-based approaches to modeling and control of particle accelerators are presented. The context for this work is important: at the start of this work in 2012, there was little interest in AI/ML in the particle accelerator community, and many of the advances in neural networks and deep learning that enabled its present success had not yet been made at that time. As such, this work was both an exploration of possible application areas and a generator of proof-of-concept demonstrations in these areas.

43 PARTICLE ACCELERATORS↗

Innovations in underground hydrogen storage with multiphysics simulations, optimization, and monitoring: A review

Underground Hydrogen Storage (UHS) is a promising solution for large-scale energy storage and a critical component in advancing low-carbon energy system. Ensuring the safety and efficiency of UHS necessitates a comprehensive understanding of multiphysical interactions driven by cyclic pore fluid pressure fluctuations and coupled physicochemical processes. Here, this review examines the key geomechanical responses in UHS, including rock property variations under cyclic loading, fracture evolution and propagation, reservoir stress sensitivity, and fault stability. It also explores the impact of geochemical and microbial reactions on geomechanical characteristics. We provide an in-depth analysis of Thermal-Hydraulic-Mechanical-Chemical (THMC) coupled numerical simulations, highlighting their potential for future multi-scale modeling. Limitations of current machine learning (ML) approaches in addressing UHS challenges are highlighted, emphasizing the need for innovative ML-based methodologies. Operational strategies for hydrogen injection and production are reviewed, focusing on safety, efficiency, and economic viability. The necessity for multi-objective optimization (MOO) to balance storage efficiency, risk mitigation, and cost-effectiveness is also discussed. Current monitoring technologies are evaluated to ensure safe and efficient UHS operations. Finally, this review identifies critical knowledge gaps and underscores the importance of advancing geomechanical understanding under multiphysics-coupling. We highlight the need for ML-driven multiphysics theories, enhanced modeling techniques, and robust optimization strategies to improve UHS performance. This study serves as a comprehensive reference for future research and the large-scale implementation of UHS systems.

25 ENERGY STORAGE↗

Data-driven learning of nonlocal physics from high-fidelity synthetic data

A key challenge to nonlocal models is the analytical complexity of deriving them from first principles, and frequently their use is justified a posteriori. Here, we extract nonlocal models from data, circumventing these challenges and providing data-driven justification for the resulting model form. Extracting data-driven surrogates is a major challenge for machine learning (ML) approaches, due to nonlinearities and lack of convexity — it is particularly challenging to extract surrogates which are provably well-posed and numerically stable. Our scheme not only yields a convex optimization problem, but also allows extraction of nonlocal models whose kernels may be partially negative while maintaining well-posedness even in small-data regimes. To achieve this, based on established nonlocal theory, we embed in our algorithm sufficient conditions on the non-positive part of the kernel that guarantee well-posedness of the learnt operator. These conditions are imposed as inequality constraints to meet the requisite conditions of the nonlocal theory. We demonstrate this workflow for a range of applications, including reproduction of manufactured nonlocal kernels; numerical homogenization of Darcy flow associated with a heterogeneous periodic microstructure; nonlocal approximation to high-order local transport phenomena; and approximation of globally supported fractional diffusion operators by truncated kernels.

42 ENGINEERING↗