SEARCH · Search NASA
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.
This presentation was part of an invited talk for a panel session presented at the 2021 IEEE Power & Energy Society General Meeting, held 26-29 July 2021.
The goal of this project was to develop and use domain-aware machine learning formulations, based on the Koopman Operator (KO), for modelling multi-scale, multi-disciplinary (e.g., multi-physics), and/or multi-agent systems. The project developed these formulations for the following cases: • Systems with dynamics at two separate time scales, • Systems with a bi-level hierarchical control structure, • Systems with bi-level hierarchical control and dynamics at two separate time scales (the lower level controls operating at the faster time scale), and • Systems with n separate but interacting agents/disciplines (with/without control, respectively); the controls for each agent could include bi-level hierarchical control and dynamics at two separate time scales as described above. The project then defined a set of dynamical systems consisting of different nonlinear oscillators that could be used to test these different formulations and then subsequently learned the KO models for those systems. With the KO models, we were able to do the following: • Quantify system stability, including both long-term and transient behavior, • Quantify the effects of feedbacks between the different time scales and agents/disciplines in terms of those feedbacks’ effects on system stability, • Replace a standard Proportional-Integral (PI) control in the hierarchical control structure with a KO-based Linear-Quadratic Regular (LQR), a form of optimal control, • Calculate optimal supervisory control policies a) with and without time scale separated dynamics at the lower level control levels and b) with both PI and KO-based LQR lower level control policies, and • Calculate dynamic Nash equilibria for multi-agent systems where each agent makes its own control decisions.
Navigating through the complex tapestry of technological advancements, "Response Strategy for Hyperparameter attacks of Digital Twin Machine Learning Model in Nuclear Power Plants" stands at the intersection of cybersecurity and nuclear power plant operations, embarking on a journey through the intricacies of securing digital twins against malicious cyber activities. As nuclear power plants progressively integrate digital twin technology and machine learning models to optimize operations and ensure system reliability, they inadvertently expose themselves to a new spectrum of vulnerabilities, notably in the realm of hyperparameter attacks. Hyperparameters, integral in machine learning model tuning and optimal performance of digital twins, have emerged as a target for adversaries aiming to destabilize the predictive capabilities and therefore, the operational accuracy of these digital entities within critical infrastructures like nuclear plants. This paper, therefore, meticulously threads the needle through the development of a robust response strategy, poised to shield these digital reflections against calculated hyperparameter manipulations, ensuring that the digital twin can effectively and securely function as a reliable proxy for its physical counterpart. The ensuing sections delve into the orchestrated maelstrom of multi-rate time-changing intelligent coordinated hyperparameter attacks and the implementation of event-triggered predictive control, laying down a structured, predictive, and responsive framework that safeguards the nexus where the digital and physical realms of nuclear power plants coalesce. The operational integrity of digital twins in nuclear power plants depends critically on the security of machine learning hyperparameters. This study makes two different contributions. First, a decision-based idea known as a multi-rate time changing intelligent coordinated hyperparameter attack is put forth. In this attack, many hyperparameters are repeatedly changed using both random and intelligent optimal techniques by the attacker. These assaults introduce varied rates at different attack steps, compromise various amounts of hyperparameters, and improve stealth and flexibility. Second, a technique is developed for event triggered predictive control to rapidly respond to potential hyperparameter attacks. This control integrates a sliding window framework, retaining a history of previous data points and employing linear regression to predict the next data point from the current dataset. The control gain K is determined using the Lyapunov-Krasovskii method, and subsequently, an action is developed. Finally, the outcome of the simulation demonstrates the viability of the proposed method for defending nuclear power plant digital twins from hyperparameter attacks.
This paper provides a summary of the methodology, findings, lessons learned, and insights from an investigation into the feasibility of the Operational Technology Behavioral Analytics (OTBA) cybersecurity approach. The concept was evaluated with data from the National Carbon Capture Center (NCCC) – a U.S. Department of Energy (DOE) funded facility that is managed and operated by Southern Company at Alabama Power’s E. C. Gaston generating power plant in Wilsonville, Alabama. Appropriate data sources for the post-combustion carbon capture system were identified. Infrastructure was deployed to monitor, capture and archive data for the system. Critical parameters for each subsystem were identified and analyzed. Machine-learning algorithms were used to establish and characterize normal operations and subsequently identify anomalies. This effort yielded valuable insights and formed the basis of a data-centric strategy for detecting cyber-attacks along with a coordinated response philosophy. A significant takeaway is that the OTBA cybersecurity approach is quite portable; it can be applied to other critical infrastructure beyond fossil power generation.
Rapid simulation of the spatiotemporal evolution of pressure & saturation for SACROC 1. Neural operator was trained on only 153 simulation runs 2. Trained to account for heterogeneity and variations/uncertainties in engineering, fluids, and geology 3. Pressure forecast has less than 1% error 4. Saturation forecast has less than 2% error 5. Traditional simulator takes 1 hour for a single scenario, while neural operator takes less than 1 minute. Rapid simulation of the spatiotemporal evolution of pressure & saturation for IBDP 1. Transfer Learning was implemented on the SACROC-based Neural Operator that was trained on only 20 simulation runs for IBDP Site 2. SACROC and IBDP Sites have several significant differences in geology and engineering parameters. 3. Pressure forecast has less than 5 psi error 4. Saturation forecast has less than 7% error 5. Traditional simulator takes 1 hour for a single scenario, while neural operator takes less than 1 minute and only 20 simulations for training/validation.
We address the challenge of acoustic simulations in three-dimensional (3D) virtual rooms with parametric source positions, which have applications in virtual/augmented reality, game audio, and spatial computing. The wave equation can fully describe wave phenomena such as diffraction and interference. However, conventional numerical discretization methods are computationally expensive when simulating hundreds of source and receiver positions, making simulations with parametric source positions impractical. To overcome this limitation, we propose using deep operator networks to approximate linear wave-equation operators. This enables the rapid prediction of sound propagation in realistic 3D acoustic scenes with parametric source positions, achieving millisecond-scale computations. By learning a compact surrogate model, we avoid the offline calculation and storage of impulse responses for all relevant source/listener pairs. Our experiments, including various complex scene geometries, show good agreement with reference solutions, with root mean squared errors ranging from 0.02 to 0.10 Pa. Notably, our method signifies a paradigm shift as—to our knowledge—no prior machine learning approach has achieved precise predictions of complete wave fields within realistic domains.
The rampdown phase of a tokamak pulse is difficult to simulate and often exacerbates multiple plasma instabilities. To reduce the risk of disrupting operations, we leverage advances in Scientific Machine Learning (SciML) to combine physics with data-driven models, developing a neural state-space model (NSSM) that predicts plasma dynamics during Tokamak à Configuration Variable (TCV) rampdowns. The NSSM efficiently learns dynamics from a modest dataset of 311 pulses with only five pulses in a reactor-relevant high-performance regime. The NSSM is parallelized across uncertainties, and reinforcement learning (RL) is applied to design trajectories that avoid instability limits. High-performance experiments at TCV show statistically significant improvements in relevant metrics. A predict-first experiment, increasing plasma current by 20% from baseline, demonstrates the NSSM’s ability to make small extrapolations. The developed approach paves the way for designing tokamak controls with robustness to considerable uncertainty and demonstrates the relevance of SciML for fusion experiments.
Accelerators increasingly use artificial intelligence (AI) and machine learning (ML) software and workflows for a variety of tasks, from optimization to fault detection and recovery. Efficient and sustainable application of these technologies necessitates specialized and facility-specific infrastructure commitments. Accelerator facilities also introduce unique radiation and security hazards, placing additional demands on operational infrastructure. These needs further escalate the prioritization of effective collaboration models and associated funding mechanisms and legal frameworks.
The development of novel instrumentation requires an iterative cycle with three stages: design, prototyping, and testing. Recent advancements in simulation and nanofabrication techniques have significantly accelerated the design and prototyping phases. Nonetheless, detector characterization continues to be a major bottleneck in device development. During the testing phase, a significant time investment is required to characterize the device in different operating conditions and find optimal operating parameters. The total effort spent on characterization and parameter optimization can occupy a year or more of an expert's time. In this work, we present a novel technique for automated sensor calibration that aims to accelerate the testing stage of the development cycle. This technique leverages closed-loop Bayesian optimization (BO), using real-time measurements to guide parameter selection and identify optimal operating states. We demonstrate the method with a novel low-noise CCD, showing that the machine learning-driven tool can efficiently characterize and optimize operation of the sensor in a couple of days without supervision of a device expert.
Precision horticulture is evolving due to scalable sensor deployment and machine learning (ML) integration. These advancements boost the operational efficiency of individual farms, balancing the benefits of analytics with autonomy requirements. However, given concerns that affect wide geographic regions (e.g., climate change), there is a need to apply models that span farms. Federated learning (FL) has emerged as a potential solution. FL enables decentralized ML across different farms without sharing private data. Traditional FL assumes simple two-tier network topologies and, thus, falls short of operating on more complex networks found in real-world agricultural scenarios. Networks vary across crops and farms and encompass various sensor data modes, extending across jurisdictions. New hierarchical FL (HFL) approaches are needed for more efficient and context-sensitive model sharing, accommodating regulations across multiple jurisdictions. Here, we present the RuralAI architecture deployment for tomato crop monitoring, featuring sensor field units for soil, crop, and weather data collection. HFL with personalization is used to offer localized and adaptive insights. Model management, aggregation, and transfers are facilitated via a flexible approach, enabling seamless communication between local devices, edge nodes, and the cloud.
This project explored coupling modeling and analysis methods from multiple domains to address complex hybrid (cyber and physical) attacks on mission critical infrastructure. Robust methods to integrate these complex systems are necessary to enable large trade-space exploration including dynamic and evolving cyber threats and mitigations. Reinforcement learning employing deep neural networks, as in the AlphaGo Zero solution, was used to identify "best" (or approximately optimal) resilience strategies for operation of a cyber/physical grid model. A prototype platform was developed and the machine learning (ML) algorithm was made to play itself in a game of 'Hurt the Grid'. This proof of concept shows that machine learning optimization can help us understand and control complex, multi-dimensional grid space. A simple, yet high-fidelity model proves that the data have spatial correlation which is necessary for any optimization or control. Our prototype analysis showed that the reinforcement learning successfully improved adversary and defender knowledge to manipulate the grid. When expanded to more representative models, this exact type of machine learning will inform grid operations and defense - supporting mitigation development to defend the grid from complex cyber attacks! This same research can be expanded to similar complex domains.
Implementing fast and high-fidelity quantum operations using open-loop quantum optimal control relies on having an accurate model of the quantum dynamics. Any deviations between this model and the complete dynamics of the device, such as the presence of spurious modes or pulse distortions, can degrade the performance of optimal controls in practice. Here, we propose an experimentally simple approach to realize optimal quantum controls tailored to the device parameters and environment while specifically characterizing this quantum system. Concretely, we use physics-inspired machine learning to infer an accurate model of the dynamics from experimentally available data and then optimize our experimental controls on this trained model. We show the power and feasibility of this approach by optimizing arbitrary single-qubit operations in detailed numerical simulations of a superconducting transmon qubit. Furthermore, we demonstrate that this framework produces an accurate description of the device dynamics under arbitrary controls, together with the precise pulses achieving arbitrary single-qubit gates with a high fidelity of ∼99.99%.
Not Available
Although a nuclear reactor is a hostile environment for sensing and electrical communications, the reactor core is amenable to acoustic communication. An acoustic measurement infrastructure (AMI) has been installed at the Advanced Test Reactor (ATR) nozzle trench area to record acoustic signals that has the ability to capture different operating regime of the reactor. This AMI includes ATR in-pile structural components, coolant, acoustic receivers, primary coolant pumps (PCP) as signal sources, a data acquisition system, and signal-processing algorithms, enabling real-time. This report will discusses development of recursive Fast Fourier Transform approach to process in real-time acoustic signals, application of short time Fast Fourier Transform to the ATR brush data to understand the vibration level and to develop spectrograms for different primary coolant pump combinations. The combination of primary coolant pumps for normal and power axial locator mechanism of ATR are different and generates different signatures. These acoustic signatures were used to develop machine learning approaches to automatically classify different operating regimes. This lay the foundation for predictive analytic framework that can be leverage by ATR to optimize their operation and maintenance. The path forward involves continued engagement with ATR and expanded implementation of AMI and predictive framework at ATR and other facilities within INL and at other experimental reactors.
During the 2022 Partner Forum at the National Renewable Energy Laboratory (NREL), stakeholders and experts from across aviation came to the whiteboard to outline common goals, align interests, understand technologies, and discuss steps for safely and seamlessly decarbonizing aviation. Attendees identified challenges, strategies, and needed research on aviation fuels, infrastructure, and aircraft. Initial insights, strategies, and high-level suggestions from an airport ground infrastructure and operations workshop are summarized below, including potential barriers and synergies for achieving bold decarbonization goals.
Residential buildings with the ability to monitor and control their net-load (sum of load and generation) can provide valuable flexibility to power grid operators. We present a novel multiclass nonintrusive load monitoring (NILM) approach that enables effective net-load monitoring capabilities at high-frequency with minimal additional equipment and cost. The proposed machine learning based solution provides accurate multiclass state predictions while operating at a faster timescale (able to provide a prediction for each 60- Hz ac cycle used in US power grid) without relying on event-detection techniques. We also introduce an innovative hybrid classification-regression method that allows for the prediction of not only load on/off states but also individual load operating power levels. A test bed with eight residential appliances is used for validating the NILM approach. Results show that the overall method has high accuracy, good scaling and generalization properties.
Operating particle accelerators involves optimizing multiple goals simultaneously, which can be challenging due to trade-offs among objectives. While evolutionary algorithms like the genetic algorithm (GA) have been used for various Multi-Objective Optimization (MOO) tasks, they are not inherently suited for complex control problems. This talk highlights two variations of Reinforcement Learning (RL) for concurrently optimizing heat load and trip rates at the Continuous Electron Beam Accelerator Facility (CEBAF). The problem involves strict constraints on individual states, actions, and overall energy requirements of the beam. First, this talk highlights how differentiability can be harnessed through a Deep Differentiable Reinforcement Learning (DDRL) approach to address MOO issues within particle accelerators. We examine the DDRL method alongside Model Free Reinforcement Learning (MFRL), GA, and Bayesian Optimization (BO). The performance of these methods is assessed by generating a Pareto-front for two objectives. Our findings indicate that DDRL excels in handling high-dimensional problems more effectively than MFRL, BO, and GA. Next, we will show integration of explainable physics-based constraints into RL algorithms to enhance trans- parency and trust in decision-making processes by enabling users to verify that agents adhere to established physical principles. This surrogate function can be modeled using neural networks or sparse dictionary mod- els. By examining the mathematical form of the learned constraint function, we are able to confirm the agent has learned to use the established physics of each environment provided but the surrogate model. In addi- tion, we find that the introduction of a mathematical functional dictionary based surrogate model enables our reinforcement learning algorithms to reliably converge for difficult high-dimensional accelerator controls environments.
This study focuses on underground natural gas storage operations and CO 2 geologic storage in saline-bearing formations; both individually and in relation to each other. This study is the first of three that evaluate analog industries of CO 2 storage (second focuses on deep well waste disposal using U.S. EPA UIC Class I wells and third on CO 2 EOR). Several pieces of information can be found throughout the study including overview of underground natural gas storage history and side-by-side comparisons of major synergistic features between underground natural gas storage and CO 2 storage in saline-bearing formations. Best practices and lessons learned from underground natural gas storage operations can provide perspective from which future CO 2 storage R&D pursuits and field projects can benefit. A summary document providing a high-level overview and key takeaways of this study is available on NETL’s website under the Collection Name: NETL’s Analog Studies to Geologic Storage of CO 2 .