Search NASA⌕ Search

SEARCH · Search NASA

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 1,801 records · Page 100

An Informatics Approach to Demand Response Optimization in Smart Grids

Power utilities are increasingly rolling out “smart” grids with the ability to track consumer power usage in near real-time using smart meters that enable bidirectional communication. However, the true value of smart grids is unlocked only when the veritable explosion of data that will become available is ingested, processed, analyzed and translated into meaningful decisions. These include the ability to forecast electricity demand, respond to peak load events, and improve sustainable use of energy by consumers, and are made possible by energy informatics. Information and software system techniques for a smarter power grid include pattern mining and machine learning over complex events and integrated semantic information, distributed stream processing for low latency response,Cloud platforms for scalable operations and privacy policies to mitigate information leakage in an information rich environment. Such an informatics approach is being used in the DoE sponsored Los Angeles Smart Grid Demonstration Project, and the resulting software architecture will lead to an agile and adaptive Los Angeles Smart Grid.

Simmhan, Yogesh↗

Learning-based Anticipatory Control of Microreactors

Learning-based anticipatory control advances the level of autonomy of microreactor control systems, a key consideration for the unattended operation of fission batteries. In this work, an anticipatory control system is shown to provide accurate load-following for a microreactor by utilizing a model predictive control framework.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Configuration of an optical switch fabric using machine learning

An optical switch fabric comprises two or more optical switch elements. The optical switch elements are configured in a topology. A switch control has a plurality of bias control signals. The switch control can address one or more of the optical switch elements and can apply one of the bias control signals to bias of the addressed optical switch element to establish a switch setting. The topology and switch settings determine how each of one of the inputs is connected to each of one of the outputs of the optical switch fabric. The switch settings are determined by a machine learning process which includes a model creation. The model can be made to adapt dynamically during optical switch fabric operation.

Dupuis, Nicolas↗

Preservation of Records, Knowledge and Memory Across Generations. An emerging Multidisciplinary Work Area and an NEA Project - 12218

Disposal in engineered facilities built in stable, deep geological formations is the reference means for permanently isolating long-lived radioactive waste from the human biosphere. This management method is designed to be intrinsically safe and final, i.e. not dependent on human presence and intervention in order to fulfil its safety goal. There is however no intention to forgo, at any time, knowledge and awareness either of the repository or of the waste that it contains. The preservation of Records, Knowledge and Memory (RK and M) is seen as an integral part of radioactive waste management, supporting lengthy and complex socio-technical processes across pre-operational, operational and post-operational lifetimes. Long-term preservation of RK and M is an emerging multidisciplinary work area in which much learning is expected over the coming years. Novel methods are being sought that are least vulnerable to both natural degradation and to changes in socio-economic conditions. Progress has been made in individual countries, but there is a need to internationalise the thinking, compare approaches, investigate potential solutions and share decisions. This is the task of the NEA RK and M project. A major outcome of the project will be a 'menu-driven document' that will allow people to identify the main elements of a strategic action plan for RK and M preservation. In sum, the preservation of RK and M is a unprecedented task in which technical, scientific and social information is interwoven and needs to be developed and preserved across generations and across specialist boundaries. Important studies have been undertaken in the past decades to explore a variety of approaches to preserving RK and M across different timescales, including archives and markers. The work of the past in this area is useful, but innovative thinking is also needed. Seen from today's perspective, very little work is available on for example the contextualization of data for later use; on the systematic identification of mechanisms for RK and M transfer; on implementing a culture of RK and M-keeping in organisations; and on creating cultural links between the waste disposals and the siting communities. Moreover, international cooperation is recognised as being crucial in providing shared means and meanings for memory transmission over longer timescales. International cooperation has also been identified as a catalyst to ensure that a wide range of approaches and experiences is considered, thus potentially reducing uncertainty related to variations in approach. Overall, multiple approaches, requiring active and less active care, need to be considered from the start of the radioactive waste management programme and refined in the course of time. The RK and M project members want to further investigate and deliver support to fulfil this task. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Stochastic Virtual Battery Modeling of Uncertain Electrical Loads using Variational Autoencoder

Effective utilization of flexible loads for grid services, while satisfying end-user preferences and constraints, requires an accurate estimation of the aggregated predictive flexibility offered by the electrical loads. Recently, there have been efforts to quantify the predictive flexibility of thermostatic loads (e.g. residential air-conditioners, electric water-heaters) using the notion of virtual battery (VB), whose state evolution is governed by a first order dynamics including self-dissipation rate, and power and energy capacities. Identifying the VB model parameters for a collection of thermostatic loads, however, is challenging primarily due to uncertainties and lack of information regarding the end-user behavior, underlying device models and parameters. In this paper, we propose a \textit{variational autoencoder}-based deep learning algorithm to identify the parameters of the VB model. Using available sensors and meters data, the proposed algorithm generates not only point estimates of the VB parameters, but also confidence intervals around those values. Effectiveness of the proposed frameworks is demonstrated on a collection of electric water-heater loads, whose operation is driven by uncertain water usage profiles.

virtual battery, deep learning algorithms↗

A novel machine learning-based optimization algorithm (ActivO) for accelerating simulation-driven engine design

A novel design optimization approach (ActivO) that employs an ensemble of machine learning algorithms is presented. The proposed approach is a surrogate-based scheme, where the predictions of a weak leaner and a strong learner are utilized within an active learning loop. The weak learner is used to identify promising regions within the design space to explore, while the strong learner is used to determine the exact location of the optimum within promising regions. For each design iteration, exploration is done by randomly selecting evaluation points within regions where the weak learner-predicted fitness is high. The global optimum obtained by using the strong learner as a surrogate is also evaluated to enable rapid convergence once the most promising region has been identified. First, the performance of ActivO was compared against five other optimizers on a cosine mixture function with 25 local optima and one global optimum. In the second problem, the objective was to minimize indicated specific fuel consumption of a compression-ignition internal combustion (IC) engine while adhering to desired constraints associated with in-cylinder pressure and emissions. In this work, the efficacy of the proposed approach is compared to that of a genetic algorithm, which is widely used within the internal combustion engine community for engine optimization, showing that ActivO reduces the number of function evaluations needed to reach the global optimum, and thereby time-to-design by 80%. Furthermore, the optimization of engine design parameters leads to savings of around 1.9% in energy consumption, while maintaining operability and acceptable pollutant emissions.

97 MATHEMATICS AND COMPUTING↗

Generic and ML Workloads in an HPC Datacenter: Node Energy, Job Failures, and Node-Job Analysis

HPC datacenters offer a backbone to the modern digital society. Increasingly, they run Machine Learning (ML) jobs next to generic, compute-intensive workloads, supporting science, business, and other decision-making processes. However, understanding how ML jobs impact the operation of HPC datacenters, relative to generic jobs, remains desirable but understudied. In this work, we leverage long-term operational data, collected from a national-scale production HPC datacenter, and statistically compare how ML and generic jobs can impact the performance, failures, resource utilization, and energy consumption of HPC datacenters. Our study provides key insights, e.g., ML-related power usage causes GPU nodes to run into temperature limitations, median/mean runtime and failure rates are higher for ML jobs than for generic jobs, both ML and generic jobs exhibit highly variable arrival processes and resource demands, significant amounts of energy are spent on unsuccessfully terminating jobs, and concurrent jobs tend to terminate in the same state. We open-source our cleaned-up data traces on Zenodo (https://doi. org/10.5281/zenodo.13685426), and provide our analysis toolkit as software hosted on GitHub (https://github.com/atlarge-research/2024-icpads-hpc-workload-characterization). This study offers multiple benefits for data center administrators, who can improve operational efficiency, and for researchers, who can further improve system designs, scheduling techniques, etc.

crossanalysis↗

Scalability analysis of heavy-duty gas turbines using data-driven machine learning

With the increasing integration of variable renewable energy sources into power systems, the role of flexible power generation technologies like gas turbines (GT) in rapid grid balancing remains crucial. This sustained importance underscores the need for scaled and precise modeling of GT to ensure effective integration within evolving energy frameworks. While physics-driven GT models integrate thermodynamics, fluid dynamics, and combustion principles, they often rely on approximate mathematical representations to accommodate scaling that may not capture the actual complex dynamics for GTs and inertial effects associated to GTs with different ratings. In this study, a data-driven model is proposed using machine learning (ML) techniques to conduct GT scalability analysis and performance evaluation with high accuracy. The ML model, trained on data from various operating conditions and performance parameters, aims to uncover intricate relationships and patterns, resembling GT characteristics at different scales (ratings). The model is developed to capture complex system interaction and to adapt to changing operational scenarios at different capacities, providing valuable insights of power system dynamics. In this study, the real-time digital simulator platform was employed to generate training data for the ML model and assess its dynamic characteristics. The ultimate objective was to develop a detailed modeling framework based on governing equations and data-driven ML capable of predicting key performance indicators, in thermal systems such as GTs, including power output, speed, fuel consumption, and exhaust temperature under diverse operating conditions at different scales. The developed ML framework demonstrated high accuracy, with mean relative errors for GT power prediction, reference speed, exhaust temperature, and compressor pressure ratio (CPR) parameters consistently below 0.1% across typical load fluctuation scenarios. Maximum deviations were limited to approximately 0.5 K for exhaust temperature and 0.009 for CPR, underscoring the model’s ability to replicating dynamic GT behavior with high precision. The adaptability of the ML model enables its application across diverse operational conditions and its extension to other thermal systems. By leveraging advanced ML techniques, this study presents a robust and scalable modeling framework that enhances GT simulation precision, facilitating improved integration into evolving power systems.

24 POWER TRANSMISSION AND DISTRIBUTION↗

SMART Deliverable 6.1.2a: Application of the ORION tool to the IBDP Carbon Storage Site

Forecasting and managing potential induced seismic activity is one of the challenges facing commercialscale geologic carbon sequestration (GCS), as well as other geologic energy extraction and byproduct disposal technologies. Historically, the process to develop robust, science-based forecasts of induced seismicity has required an integrated effort from experts in seismology, geomechanics, and reservoir engineering to manage data, develop and evaluate models of subsurface processes, and to calibrate and interpret the results from a range of models to understand site behavior relative to prescribed standards and in the context of uncertainty in geologic characterization data, forecasting models, and operational scenario uncertainty. The Operational Forecasting of Induced Seismicity (ORION) toolkit is an open-source, observation-based forecasting toolkit that is being co-developed by two U.S. DOE-funded initiatives: the National Risk Assessment Partnership (NRAP) and the Scienceinformed Machine Learning for Accelerating Real Time Decisions in Subsurface Applications (SMART) Initiative. ORION is designed to provide functionality to support decision making about seismic hazard analysis and risk management for GCS stakeholders ranging from the public to site operators to expert seismologists. The tool, which is written as open-source code in the Python programming language, is composed of a desktop graphical user interface (GUI) and an underlying forecasting engine. The forecasting engine uses available reservoir properties, well and fluid injection scenario details, and observed seismic catalog data as inputs to produce a set of temporal and spatio-temporal seismic forecasts.

58 GEOSCIENCES↗

Capturing and Using Knowledge about the Use of Visualization Toolkits

When constructing visualization pipelines using toolkits such as Visualization Toolkit (VTK) and Generic Mapping Tools (GMT), developers must understand (1) what toolkit operators will transform their data from its raw state to some required view state and (2) what viewers are available to present the generated view. Traditionally, developers learn about how to construct visualization pipelines by reading documentation and inspecting code examples, which can be costly in terms of the time and effort expended. Once an initial pipeline is constructed, developers may still have to undergo a trial and error process before a satisfactory visualization is generated. This paper presents the Visualization Knowledge Project (VisKo) that is built on a knowledge base of visualization toolkit operators and how they can be piped together to form visualization pipelines. Developers may now rely on VisKo to guide them when constructing visualization pipelines and in some cases, when VisKo has complete knowledge about some set of operators (i.e., sequencing and parameter settings), automatically generate a fully functional visualization pipeline.

Chemical Imaging↗

BOCC Blueprint

The Pacific Northwest National Laboratory (PNNL) Facility and Operations (F&O) Division strives to manage performance, to assure long-term health and sustainability for the organization and our staff through innovation and learning. The F&O organization is committed to achieving five strategic outcomes over the next several years. The strategic outcomes are aligned with the Department of Energy (DOE) and PNNL mission and vision. One key outcome important to enabling our vision is to integrate the smartest facilities, connected to the smartest people, by the smartest systems. We call this our SMART3 Initiative. The F&O SMART3 Initiative will connect human capital with intelligent systems, resulting in a best-in-class, integrated facility-operations strategy that improves productivity and optimizes operational costs. The Building Operations Control Center (BOCC) Blueprint is a key deliverable in this effort and will define near-term and long-term objectives aligned with the SMART3 Initiative. The BOCC Blueprint provides a plan to successfully position the BOCC and assure full integration into the new Systems Engineering Laboratory (SEL). The BOCC Blueprint near-term objectives will satisfy each of the three elements of the SMART3 Initiative. 1. The facilities element will be achieved by defining the minimum set of smart sensors for deployment, establishing dashboards and metrics, and initiating a Real-Time Commissioning Program (RT-Cx). 2. The people element will be achieved as F&O and the Energy and Environment Directorate (EED) collaborate, mentor, and cross-train within F&O. The collaboration effort will provide proof of concept by testing, applying and sharing information on facility systems through smart grid, facility re-tuning & audits, and DSOM fault detection and diagnostics. The mentoring and cross-training will be realized by integrating air balance operators, power operators and FMCS specialists within the BOCC. 3. The systems element will be achieved by standardizing the Facility Management Control Systems (FMCS) and Decision Support for Operations and Maintenance (DSOM) development, specifically by implementing facility dashboards and tools. The long-term objectives are to enhance reliability, reduce cost, and demonstrate the “Living Laboratory,” as defined by the way facilities are used to advance research from theoretical to practical applications through scientific collaboration among PNNL’s research and development (R&D) staff. With the BOCC Blueprint defining the objectives to meet the SMART3 Initiative, the BOCC will become a valuable PNNL core competency, and allow for additional R&D opportunities.

99 GENERAL AND MISCELLANEOUS↗

Data-driven building energy modeling with feature selection and active learning for data predictive control

Three gaps impede the development of cost-effective and accurate data-driven building energy modeling/models (DBEM) for energy forecasting and predictive control strategies. Gap 1: data bias is common in building operation data, but this topic is hardly studied in DBEM; Gap 2: high data dimensionality is common in DBEM, but a systematic and scalable methodology is lacking to solve the problem; Gap 3: the interactions between data bias and high dimensionality have not been systematically studied for DBEM and predictive control in buildings. In this work, to address the three gaps mentioned above, we develop a framework that integrates active learning and feature selection for DBEM used for whole building data predictive control (or DPC, which is a branch of model predictive control). The framework provides a systematic methodology and automatic workflow that starts with raw data from building automation systems to the establishment of data-driven energy models for DPC controllers. The developed strategies and framework are evaluated in a virtual testbed based on EnergyPlus and BCVTB. Improved performance and reduced computational complexity are observed from the DBEM built with the developed framework, as well as the DPC controller based on that DBEM, indicating the effectiveness of the developed framework.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Multi-level optimization with the koopman operator for data-driven, domain-aware, and dynamic system security

Cyber-Physical Systems (CPSs) like the power grid are critically important but also increasingly vulnerable; ensuring reliable system operation in the face of disruptions is becoming more and more challenging. Multi-Level Optimization (MLO) is a powerful way to model adversarial interactions, which naturally makes it applicable to studying CPS security. However, MLO typically does not address underlying system dynamics, and incorporating nonlinear dynamics is generally infeasible. In this paper, we show how to combine MLO with the Koopman Operator (KO) to remedy this. The KO maps nonlinear dynamics to a lifted space in which those dynamics are linear, thus making it ideal for use with MLO. Moreover, the structure of the KO also provides convenient ways to incorporate domain knowledge into the data-driven process of learning the KO representation of a given system. Here we then demonstrate the use of MLO-KO on a small example problem taken from the power grid domain, discuss the scalability and computational cost of MLO-KO, and identify future research directions for this work.

42 ENGINEERING↗

Bearing Fault Detection and Wear Estimation Using Machine Learning

Unpredicted bearings failures can be costly and dangerous. Current methods have been examining the use of machine learning algorithms to automatically detect bearing faults. Vibration specialists must react daily, as the sheer number of bearings and the machines they operate are critical and any downtime will result in loss of productivity and endanger personal. The method presented in this paper utilizes a novel approach to detect and determine bearings fault diameters. The detection algorithm uses the cubic support vector machine (SVM) for high accuracy inner and outer race fault detection. After detecting the type of fault, the data is then passed through a second layer that uses a modified K-means clustering and a linear interpolation to accurately estimate the diameter of the fault that was detected.

42 ENGINEERING↗

SNM Radiation Signature Classification Using Different Semi-Supervised Machine Learning Models

The timely detection of special nuclear material (SNM) transfers between nuclear facilities is an important monitoring objective in nuclear nonproliferation. Persistent monitoring enabled by successful detection and characterization of radiological material movements could greatly enhance the nuclear nonproliferation mission in a range of applications. Supervised machine learning can be used to signal detections when material is present if a model is trained on sufficient volumes of labeled measurements. However, the nuclear monitoring data needed to train robust machine learning models can be costly to label since radiation spectra may require strict scrutiny for characterization. Therefore, this work investigates the application of semi-supervised learning to utilize both labeled and unlabeled data. As a demonstration experiment, radiation measurements from sodium iodide (NaI) detectors are provided by the Multi-Informatics for Nuclear Operating Scenarios (MINOS) venture at Oak Ridge National Laboratory (ORNL) as sample data. Anomalous measurements are identified using a method of statistical hypothesis testing. After background estimation, an energy-dependent spectroscopic analysis is used to characterize an anomaly based on its radiation signatures. In the absence of ground-truth information, a labeling heuristic provides data necessary for training and testing machine learning models. Supervised logistic regression serves as a baseline to compare three semi-supervised machine learning models: co-training, label propagation, and a convolutional neural network (CNN). In each case, the semi-supervised models outperform logistic regression, suggesting that unlabeled data can be valuable when training and demonstrating value in semi-supervised nonproliferation implementations.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗

Machine Learning-Enabled Wearable Piezoelectric Acoustic Sensor for Real-Time Breast Abnormality Detection

In contemporary society, breast health has become a significant public health concern, particularly among women. According to statistics from the World Health Organization, both the incidence and mortality rates of breast tumors have steadily increased in recent years. Therefore, effective early-stage screening and postoperative monitoring are essential for maintaining breast health. However, conventional clinical diagnostic modalities are typically bulky, operationally complex, and unsuitable for continuous real-time monitoring, which limits their use in portable and everyday health management applications. To address these limitations, this study proposes a machine learning-integrated wearable piezoelectric sensing platform as an auxiliary tool for breast health assessment. The device consists of a PDMS matching layer embedded with flexible silver nanowires, a P(VDF-TrFE) piezoelectric layer, and a multi-channel low-noise signal acquisition circuit. It is capable of acquiring acoustic echo signals from tissue-mimicking environments and automatically evaluating signal validity using a convolutional neural network (CNN). By integrating piezoelectric sensing with deep learning-based signal analysis, the proposed system achieves a signal-to-noise ratio exceeding 70 dB and a real-time classification accuracy above 96% under controlled conditions. These results demonstrate that the platform provides a compact, portable, and intelligent approach for wearable sensing of mechanical heterogeneity and highlight its potential for future development in continuous biomedical monitoring technologies.

He, Shuaitong↗

Lessons Learned in Implementation of Radiological Investigations of Small-Scale Sites in Port Hope, Ontario - 16534

The Port Hope Area Initiative is a $1.28 billion project for the remediation of historic low-level radioactive waste resulting from the operation of Eldorado Nuclear Limited and its private-sector predecessors between 1932 and 1988 in Port Hope and Port Granby, Ontario, approximately 100 km east of Toronto. Included in the project is the investigation of all 4,800 privately and publicly owned properties in Port Hope in order to identify and remediate an estimated 375 properties containing LLRW exceeding clean-up criteria. This paper discusses the progress of the investigations, the development of clean-up criteria, methodologies used to identify historic LLRW, and lessons learned to date. This paper will be of greatest interest to future project managers of similar investigations. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

A Scalable Pipeline for Gigapixel Whole Slide Imaging Analysis on Leadership Class HPC Systems

Whole Slide Imaging (WSI) captures microscopic details of a patient's histopathological features at multiple res-olutions organized across different levels. Images produced by WSI are gigapixel-sized, and saving a single image in memory requires a few gigabytes which is scarce since a complicated model occupies tens of gigabytes. Performing a simple met-ric operation on these large images is also expensive. High-performance computing (HPC) can help us quickly analyze such large images using distributed training of complex deep learning models. One popular approach in analyzing these images is to divide a WSI image into smaller tiles (patches) and then train a simpler model with these reduced-sized but large numbers of patches. However, we need to solve three pre-processing challenges efficiently for pursuing this patch-based approach. 1) Creating small patches from a high-resolution image can result in a high number (hundreds of thousands per image) of patches. Storing and processing these images can be challenging due to a large number of I/O and arithmetic operations. To reduce I/Oand memory accesses, an optimal balance between the size and number of patches must exist to reduce I/O and memory accesses. 2) WSI images may have tiny annotated regions for cancer tissue and a significant portion with normal and fatty tissues; correct patch sampling should avoid dataset imbalance. 3) storing and retrieving many patches to and from disk storage might incur I/O latency while training a deep learning model. An efficient distributed data loader should reduce I/O latency during the training and inference steps. This paper explores these three challenges and provides empirical and algorithmic solutions deployed on the Summit supercomputer hosted at the Oak Ridge Leadership Computing Facility.

Dash, Sajal↗

Refine your search to explore more results.