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 109 records · Page 6

Enhancing the Operational Resilience of Advanced Reactors with Digital Twins by Recurrent Neural Networks

Because of a lack of operation data during abnormal and accident scenarios, along with the existence of uncertainty in the evaluation model for transient and accident analysis, the established abnormal and emergency operating procedures can be biased in characterizing the reactor states and ensuring operational resilience. To improve state awareness and ensure operational flexibility for minimizing effects on the system due to anomaly, digital twin (DT) technology is suggested to support operator's decision-making by effectively extracting and using knowledge of the current and future plant states from the knowledge base. To demonstrate DT's capability for recovering the complete states of reactors and for predicting the future reactor behaviors, this paper develops and assesses both the diagnosis and prognosis DTs in a nearly autonomous management and control system for an Experimental Breeder Reactor-II simulator during different loss-of-flow scenarios.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Improving the prediction of daily reservoir releases over the CONUS using conditioned LSTM

Reservoirs play a vital role in regulating streamflow timing and variability for hydroelectricity, flood control, water supply, irrigation, and recreation. Despite their importance, many reservoirs lack comprehensive operational guidelines, making their management complex due to conflicting operational objectives. Hence traditional policy-based reservoir models often fail to capture real-world conditions accurately and they depend on perfect streamflow predictions, which are not always available. In contrast, data-driven models like Long Short-Term Memory (LSTM) networks offer a robust alternative. This study introduces an approach that integrates reservoir characteristics—such as main use, climate, and maximum capacity—into the LSTM model to enhance reservoir release predictions. Using data from nearly 200 reservoirs in the contiguous United States (CONUS), our conditioned LSTM model (LSTM_cond) was compared with both the vanila LSTM and a traditional policy-based approach. Furthermore, our results show that while both LSTM_cond and LSTM perfoms better than the policy-based approach, LSTM_cond consistently outperforms LSTM for hydroelectric, water supply, irrigation, and recreation reservoirs. The KGE median values for LSTM_cond for out-sample reservoirs are 0.764, 0.565, 0.821, and 0.779, respectively, for the aforementioned reservoir types, which are consistently higher that the corresponding KGE values of 0.737, 0.413, 0.775, and 0.713 of LSTM, demonstrating its advantages in improving generalizability.

CONUS↗

Curating a Standardized Dataset for Statistical Biosignature Classification

In recent years, machine learning has been explored as a toolkit for planetary science and operations [Helbert, Azari]. Machine learning has been used to improve our understanding of possible biosignatures and mineral signatures to improve science return on future missions [Warren-Rhodes, Cleaves].

Biosignatures↗

Optimizing inference of segmentation on high-resolution images in MLExchange

MLExchange is a machine learning (ML) operations platform providing web user-interfaces (UIs) for data visualization and analysis pipelines at synchrotron facilities. Among these UIs is the segmentation app which helps synchrotron users utilize ML algorithms to automatically segment high-resolution scientific images with minimal manual annotation effort. In this work, we share code optimizations that significantly speed up the segmentation inference workflow of large data in short time. By optimizing the sequence of CPU-GPU data transfers and introducing CPU parallelization to key operations, we improve the per-device, per-image frame computational efficiency and observe close to 3×$$\times$$ speedup over the original segmentation inference workflow run time when utilizing a single GPU. Further adaptations enabling multi-GPU inference yield more than 40×$$\times$$ speedup with 100 GPUs compared to the optimized single GPU inference workflow. This acceleration of the segmentation inference workflow will provide MLExchange users with easy access to segmentation results with little wait time.

Lu, Shizhao↗

Online Electron Reconstruction at CLAS12

Online reconstruction plays a crucial role in monitoring and in real-time analysis of high energy and nuclear physics experiments. A vital aspect of reconstruction algorithms is particle identification, which combines information from various detector components to determine the type of particle. Electron identification is particularly significant in electro-production nuclear physics experiments like the CLAS12 spectrometer at Jefferson Laboratory as it is essential in data recording. A machine learning approach has been developed for CLAS12 experiments to reconstruct and identify electrons by combining raw signals from multiple detector components at the data acquisition level. This method achieves high electron identification purity while maintaining nearly 100% efficiency. Furthermore, the machine learning tools operate at rates exceeding data acquisition speed, enabling the real-time electron reconstruction. This advancement significantly improves online analyses and monitoring capabilities for CLAS12 experiments.

Tyson,, Richard [Thomas Jefferson National Acceler↗

Exact constraints and appropriate norms in machine-learned exchange-correlation functionals

Machine learning techniques have received growing attention as an alternative strategy for developing general-purpose density functional approximations, augmenting the historically successful approach of human-designed functionals derived to obey mathematical constraints known for the exact exchange-correlation functional. More recently, efforts have been made to reconcile the two techniques, integrating machine learning and exact-constraint satisfaction. We continue this integrated approach, designing a deep neural network that exploits the exact constraint and appropriate norm philosophy to de-orbitalize the strongly constrained and appropriately normed (SCAN) functional. The deep neural network is trained to replicate the SCAN functional from only electron density and local derivative information, avoiding the use of the orbital-dependent kinetic energy density. The performance and transferability of the machine-learned functional are demonstrated for molecular and periodic systems.

Artificial neural networks↗

Machine Learning for Daily Forecasts of Arctic Sea Ice Motion: An Attribution Assessment of Model Predictive Skill

Physics-based simulations of Arctic sea ice are highly complex, involving transport between different phases, length scales, and time scales. Resultantly, numerical simulations of sea ice dynamics have a high computational cost and model uncertainty. We employ data-driven machine learning (ML) to make predictions of sea ice motion. The ML models are built to predict present-day sea ice velocity given present-day wind velocity and previous-day sea ice concentration and velocity. Models are trained using reanalysis winds and satellite-derived sea ice properties. We compare the predictions of three different models: persistence (PS), linear regression (LR), and a convolutional neural network (CNN). We quantify the spatiotemporal variability of the correlation between observations and the statistical model predictions. Additionally, we analyze model performance in comparison to variability in properties related to ice motion (wind velocity, ice velocity, ice concentration, distance from coast, bathymetric depth) to understand the processes related to decreases in model performance. Results indicate that a CNN makes skillful predictions of daily sea ice velocity with a correlation up to 0.81 between predicted and observed sea ice velocity, while the LR and PS implementations exhibit correlations of 0.78 and 0.69, respectively. The correlation varies spatially and seasonally: lower values occur in shallow coastal regions and during times of minimum sea ice extent. LR parameter analysis indicates that wind velocity plays the largest role in predicting sea ice velocity on 1-day time scales, particularly in the central Arctic. Regions where wind velocity has the largest LR parameter are regions where the CNN has higher predictive skill than the LR.

54 ENVIRONMENTAL SCIENCES↗

Non-Intrusive Parallel-in-Time Solvers for Partial Differential Equations (Final Report)

Many time-dependent problems and simulations are often modeled using Partial Differential Equations. Traditional modeling approaches that use sequential time-stepping are reaching a bottleneck in optimizing efficiency. The Center of Applied Science and Computing at Lawrence Livermore National Laboratory extensively works on parallelizing these algorithms to leverage the increasing computational power from the growing number of processors in computer hardware. In particular, they aim to design non-intrusive algorithms that can generalize to a variety of problems and sizes without requiring additional information from or modifications on the original problems. Multigrid Reduction in Time (MGRIT) is a parallel-in-time algorithm that is designed to be non-intrusive. This project focuses on increasing the efficiency of MGRIT by approximating the coarse-grid operator using machine learning approaches as a means to find the most non-intrusive, or general, solution.

97 MATHEMATICS AND COMPUTING↗

Cyber Resilience and Social Equity: Twin Pillars of a Sustainable Energy Future

This paper examines the intersection of security and accessibility within energy systems amidst the rise of grid modernization and digitization, especially considering the regulatory changes and the imperatives of inclusive energy strategies. It addresses the dual need for secure, resilient infrastructure and a commitment to mitigate energy poverty while maintaining equitable access to energy. Amid escalating cybersecurity and physical threats, the paper advocates for sustainable energy delivery systems that ensure robust defenses without compromising the goals of reducing energy poverty and ensuring energy security. This paper identifies the pressing need for Cyber-Informed Engineering (CIE) and Secure-by-Design (SbD) principles, highlighting how these strategies can protect critical infrastructure and democratize access to secure energy, particularly for disadvantaged communities. The analysis underscores the challenges presented by the expansion of attack surfaces, interoperability requirements, and grid-edge analytics, offering innovative solutions that leverage advanced technologies and data-driven insights. Furthermore, this paper addresses the workforce development gap, emphasizing the necessity for public-private partnerships and vendor engagement in creating a skilled cybersecurity workforce. This paper has a dual focus on both the technological aspect of cybersecurity and the social dimension of equity within the context of sustainable energy development. It suggests a comprehensive examination of how these two critical elements interact and support the overarching goal of a sustainable energy future.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Characterization of Response Times based on Voice Communication and Traffic Surveillance Data

A barrier to the integration of remotely piloted aircraft operations in the U.S. National Airspace System is the latency of voice communications between the air traffic controller and the remote pilot, and the latency of communication between the aircraft and the remote pilot. The latency can be substantial especially when satellite-based beyond-radio-line-of-sight communication and relay through the aircraft are employed. This study uses voice recordings of controller-pilot communications and aircraft track data to establish a baseline of pilot readback latencies and maneuver detection delays in the current piloted operations. A machine learning pipeline was developed to parse the contents of the air traffic control clearances including the callsigns using natural language processing. After manually validating the results obtained using the pipeline, the average pilot readback latency was found to be about 0.6 seconds. The average latency between the end of maneuver (inferred from track data), initiated by the pilot in response to the clearance, and the end of clearance was found to be about 176 seconds for altitude change commands, 69 seconds for heading change commands, and 182 seconds for speed change commands. The average latency between the beginning of maneuver and the end of clearance was found to be about 17 seconds for altitude change commands, 17seconds for heading change commands, and 25 seconds for speed change commands.

controller-pilot communication, communication late↗

Characterization of Response Times Based on Voice Communication and Traffic Surveillance Data

A barrier to the integration of remotely piloted aircraft operations in the U.S. National Airspace System is the latency of voice communications between the air traffic controller and the remote pilot, and the latency of communication between the aircraft and the remote pilot. The latency can be substantial especially when satellite-based beyond-radio-line-of-sight communication and relay through the aircraft are employed. This study uses voice recordings of controller-pilot communications and aircraft track data to establish a baseline of pilot readback latencies and maneuver detection delays in the current piloted operations. A machine learning pipeline was developed to parse the contents of the air traffic control clearances including the callsigns using natural language processing. After manually validating the results obtained using the pipeline, the average pilot readback latency was found to be about 0.6 seconds. The average latency between the end of maneuver (inferred from track data), initiated by the pilot in response to the clearance, and the end of clearance was found to be about 176 seconds for altitude change commands, 69 seconds for heading change commands, and 182 seconds for speed change commands. The average latency between the beginning of maneuver and the end of clearance was found to be about 17 seconds for altitude change commands, 17seconds for heading change commands, and 25 seconds for speed change commands.

controller-pilot communication↗

Real-time avoidance of the L-mode and H-mode density limit via machine-learned stability metrics

Reliable operation of burning plasma tokamaks will require robust control strategies to avoid macroscopic instability limits such as the L-mode and H-mode density limits (LDL, HDL). In this work, we explore closed-loop avoidance of these phenomena at DIII-D using machine-learned risk metrics. Feedback control is implemented via the ‘DL Supervisor’ scheme, which regulates the chosen risk metric by reducing the density target or increasing NBI heating in real-time. Using the LDL 25 risk metric, the LDL is reproducibly suppressed. We also introduce an HDL risk metric in this study, HDL 25 , which reduces the False Positive Rate by 2x compared to the Greenwald fraction. Applying this scaling to a plasma current ramp-down, we successfully avoid an HDL-driven H/L back-transition. These experiments constitute the first demonstration of real-time DL avoidance using machine-learned risk metrics. These instability metrics outline a path to safer high-density operation, more reliable ramp-down scenarios, and improved off-normal control for next-step devices such as ITER and SPARC.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

MSL Telecom Automated Anomaly Detection

The Mars Science Laboratory (MSL) Telecom Operations Team at the Jet Propulsion Laboratory (JPL) has implemented a machine learning system in order to automate the anomaly detection process as a part of daily operations. Machine learning enables reliable detection of anomalies in Telecom-related telemetry and automated reporting of Telecom subsystem status, resulting in an 90% reduction in team workload and improved anomaly detection reliability. At present, machine learning methods are used to detect: 1. Anomalous long-term trends in telemetry data 2. Anomalous time-domain evolution of telemetry values Both types of anomalies pose their own unique challenges that are addressed in different ways. In the first case, long term trending of daily minima, maximum, and mean telemetry values in temperatures, currents, voltages, and radio frequency (RF) power levels is used in addition to hard threshold safety checks to look for changes in long-term equipment health and performance. Long-term trending methods allow for ordinary seasonal variations in these quantities caused by temperature changes over the course of the Martian year while allowing operators to determine whether current performance remains in line with historical values from previous years. Changes in long-term trends can provide important insights into the health and status of the rover's on-board systems as well as valuable early warning if subtle degradation begins to take hold. But while trending of daily statistics is valuable, it does not detect anomalies in the short-term time evolution of data over the course of minutes or hours during a day, and this task is handled with short-term shape analysis. Principal components analysis (PCA) has been found to provide robust detection of short-term anomalies, and several examples of the use of PCA to detect actual anomalous events will be provided here. In using PCA, we use both the percentage of explained variance and also a log likelihood test on the PCA expansion coefficients to flag telemetry data for human review. Previous work in the field of spacecraft anomaly detection includes [1] for MSL and [2] for some other JPL missions.

Mukai, Ryan↗