Search NASA⌕ Search

SEARCH · Search NASA

Results for “Operational adaptability”

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 793 records · Page 44

An Adaptive Newton-Based Free-Boundary Grad–Shafranov Solver

Equilibria in magnetic confinement devices result from force balancing between the Lorentz force and the plasma pressure gradient. In an axisymmetric configuration like a tokamak, such an equilibrium is described by an elliptic equation for the poloidal magnetic flux, commonly known as the Grad–Shafranov equation. It is challenging to develop a scalable and accurate free-boundary Grad–Shafranov solver, since it is a fully nonlinear optimization problem that simultaneously solves for the magnetic field coil current outside the plasma to control the plasma shape. In this work, we develop a Newton-based free-boundary Grad–Shafranov solver using adaptive finite elements and preconditioning strategies. The free-boundary interaction leads to the evaluation of a domain-dependent nonlinear form of which its contribution to the Jacobian matrix is achieved through shape calculus. The optimization problem aims to minimize the distance between the plasma boundary and specified control points while satisfying two nontrivial constraints, which correspond to the nonlinear finite element discretization of the Grad–Shafranov equation and a constraint on the total plasma current involving a nonlocal coupling term. The linear system is solved by a block factorization, and AMG is called for subblock elliptic operators. The unique contributions of this work include the treatment of a global constraint, preconditioning strategies, nonlocal reformulation, and the implementation of adaptive finite elements. Furthermore, it is found that the resulting Newton solver is robust, successfully reducing the nonlinear residual to 1e-6 and lower in a small handful of iterations while addressing the challenging case to find a Taylor state equilibrium where conventional Picard-based solvers fail to converge.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Development of efficient time-evolution method based on three-term recurrence relation

The advantage of the real-time (RT) propagation method is a direct solution of the time-dependent Schrödinger equation which describes frequency properties as well as all dynamics of a molecular system composed of electrons and nuclei in quantum physics and chemistry. Its applications have been limited by computational feasibility, as the evaluation of the time-evolution operator is computationally demanding. In this article, a new efficient time-evolution method based on the three-term recurrence relation (3TRR) was proposed to reduce the time-consuming numerical procedure. The basic formula of this approach was derived by introducing a transformation of the operator using the arcsine function. Since this operator transformation causes transformation of time, we derived the relation between original and transformed time. The formula was adapted to assess the performance of the RT time-dependent Hartree-Fock (RT-TDHF) method and the time-dependent density functional theory. Compared to the commonly used fourth-order Runge-Kutta method, our new approach decreased computational time of the RT-TDHF calculation by about factor of four, showing the 3TRR formula to be an efficient time-evolution method for reducing computational cost.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Fast Numerical Solution of the Plasma Response Matrix for Real-time Ideal MHD Control

To help effectuate near real-time feedback control of ideal MHD instabilities in tokamak geometries, a parallelized version of A.H. Glasser’s DCON (Direct Criterion of Newcomb) code is developed. To motivate the numerical implementation, we first solve DCON’s δW formulation with a Hamilton-Jacobi theory, elucidating analytical and numerical features of the ideal MHD stability problem. The plasma response matrix is demonstrated to be the solution of an ideal MHD Riccati equation. We then describe our adaptation of DCON with numerical methods natural to solutions of the Riccati equation, parallelizing it to enable its operation in near real-time. We replace DCON’s serial integration of perturbed modes—which satisfy a singular Euler- Lagrange equation—with a domain-decomposed integration of state transition matrices. Output is shown to match results from DCON with high accuracy, and with computation time < 1s. Such computational speed may enable active feedback ideal MHD stability control, especially in plasmas whose ideal MHD equilibria evolve with inductive timescale $\tau$ ≳ 1s—as in ITER. Further potential applications of this theory are discussed.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Reducing Communication Overhead in Federated Learning for Network Anomaly Detection with Adaptive Client Selection

Communication overhead in federated learning (FL) poses a significant challenge for network anomaly detection systems, where the myriad of client configurations and network conditions can severely impact system efficiency and detection accuracy. While existing approaches attempt to address this through individual optimization techniques, they often fail to maintain the delicate balance between reduced overhead and detection performance. This paper presents an adaptive FL framework that dynamically combines batch size optimization, client selection, and asynchronous updates to achieve efficient anomaly detection. Through extensive profiling and experimental analysis on two distinct datasets-UNSW-NBIS for general network traffic and ROAD for automotive networks-our framework reduces communication overhead by 97.6%; (from 700.0s to 16.8s) compared to synchronous baseline approaches while maintaining comparable detection accuracy (95.10%; vs. 95.12%;). Statistical validation using Mann-Whitney U test confirms significant improvements (p < 0.05) over existing FL approaches across both datasets, demonstrating the framework's adaptability to different network security contexts. Detailed profiling analysis reveals the efficiency gains through dramatic reductions in GPU operations and memory transfers while maintaining robust detection performance under varying client conditions.

Marfo, William [University of Texas at El Paso]↗

A Length Adaptive Algorithm-Hardware Co-design of Transformer on FPGA Through Sparse Attention and Dynamic Pipelining

Transformers are considered one of the most important deep learning models since 2018, in part because it establishes state-of-the-art (SOTA) records and could potentially replace existing Deep Neural Networks (DNNs). Despite the remarkable triumphs, the prolonged turnaround time of Transformer models is a widely recognized roadblock. The variety of sequence lengths imposes additional computing overhead where inputs need to be zero-padded to the maximum sentence length in the batch to accommodate the parallel computing platforms. This paper targets the field-programmable gate array (FPGA) and proposes a coherent sequence length adaptive algorithm–hardware co-design for Transformer acceleration. Particularly, we develop a hardware-friendly sparse attention operator and a length-aware hardware resource scheduling algorithm. The proposed sparse attention operator brings the complexity of attention-based models down to linear complexity and alleviates the off-chip memory traffic. The proposed length-aware resource hardware scheduling algorithm dynamically allocates the hardware resources to fill up the pipeline slots and eliminates bubbles for NLP tasks. Experiments show that our design has very small accuracy loss and has 80.2 × and 2.6 × speedup compared to CPU and GPU implementation, and 4 × higher energy efficiency than state-of-the-art GPU accelerator optimized via CUBLAS GEMM.

Peng, Hongwu↗

OpenFlow arbitrated programmable network channels for managing quantum metadata

Quantum networks must classically exchange complex metadata between devices in order to carry out information for protocols such as teleportation, super-dense coding, and quantum key distribution. Demonstrating the integration of these new communication methods with existing network protocols, channels, and data forwarding mechanisms remains an open challenge. Software-defined networking (SDN) offers robust and flexible strategies for managing diverse network devices and uses. We adapt the principles of SDN to the deployment of quantum networks, which are composed from unique devices that operate according to the laws of quantum mechanics. We show how quantum metadata can be managed within a software-defined network using the OpenFlow protocol, and we describe how OpenFlow management of classical optical channels is compatible with emerging quantum communication protocols. We next give an example specification of the metadata needed to manage and control quantum physical layer (QPHY) behavior and we extend the OpenFlow interface to accommodate this quantum metadata. Here, we conclude by discussing near-term experimental efforts that can realize SDN’s principles for quantum communication.

97 MATHEMATICS AND COMPUTING↗

WindWaveFloat (WWF): Final Scientific Report

Principle Power Inc. and National Renewable Energy Lab (NREL) have completed a contract to assess the technical and economic feasibility of integrating wave energy converters into the WindFloat, resulting in a new concept called the WindWaveFloat (WWF). The concentration of several devices on one platform could offer a potential for both economic and operational advantages. Wind and wave energy converters can share the electrical cable and power transfer equipment to transport the electricity to shore. Access to multiple generation devices could be simplified, resulting in cost saving at the operational level. Overall capital costs may also be reduced, provided that the design of the foundation can be adapted to multiple devices with minimum modifications. Finally, the WindWaveFloat confers the ability to increase energy production from individual floating support structures, potentially leading to a reduction in levelized energy costs, an increase in the overall capacity factor, and greater stability of the electrical power delivered to the grid. The research conducted under this grant investigated the integration of several wave energy device types into the WindFloat platform. Several of the resulting system designs demonstrated technical feasibility, but the size and design constraints of the wave energy converters (technical and economic) make the WindWaveFloat concept economically unfeasible at this time. Not enough additional generation could be produced to make the additional expense associated with wave energy conversion integration into the WindFloat worthwhile.

16 TIDAL AND WAVE POWER↗

Collaborative Proposal: Transforming How Climate System Models are Used: A Global, Multi-Resolution Approach

Despite the great interest in regional modeling for both weather and climate applications, regional modeling is not yet at the stage that it can be used routinely and effectively for climate modeling of the ocean. The overarching goal of this project is to transform how climate models are used by developing and implementing a robust, efficient, and accurate global approach to regional ocean modeling. To achieve this goal, we will use theoretical and computational means to resolve several basic modeling and algorithmic issues.

54 ENVIRONMENTAL SCIENCES↗

Machine Learning and Understanding for Intelligent Extreme Scale Scientific Computing and Discovery. DOE Workshop Report, January 7–9, 2015, Rockville, MD

Large-scale parallel simulations and data analysis drive scientific discovery across many disciplines. To drive larger and more detailed simulations, and deal with larger data volumes, exascale machines capable of 1018 operations per second are expected within the next five to ten years. However, the complexity of developing and adapting modern simulation codes for these architectures is increasing rapidly. Workloads on exascale machines will be billion-way parallel. Exploiting the full capability of the machine will require carefully assigning application tasks to cores, accelerators, deep memories, and other heterogeneous compute resources, while simultaneously optimizing for time to solution, data movement, power, and resilience. Optimizations that improve performance on one machine may slow down another. Worse, applications themselves are dauntingly complex. Production simulations comprise millions of lines of code and use sophisticated, adaptive algorithms whose performance is input-dependent. Complex workflows can couple multi-physics simulation with data preprocessing and post-processing modules.

97 MATHEMATICS AND COMPUTING↗

Machine Learning Algorithms for Matching Theories, Simulations, and Observations in Cosmology (Final Project)

The search for simple and elegant equations to explain and predict observable natural phenomena has been a driving force in scientific theory for several centuries. The great successes of Newton's laws, Maxwell's equations, the Schrodinger equation, and the theory of General Relativity have encouraged this perspective, and perhaps a beautiful equation comprising a “theory of everything” remains to be discovered. But now scientists are grappling with systems of ever-increasing complexity, fraught with nonlinearity, and with the quantities of interest often only indirectly observed. Nature may resist a simple description, and the most important discoveries of the next century may be complex theories with countless variables and parameters. The era of big data opens up a promising new approach to scientific discovery in this setting. The predictions of modern theories, even if complex or nonlinear, can be examined through detailed and computationally-intensive simulations that take days or weeks to run and fill petabytes of disk space. The upcoming generation of scientific instruments will provide petabytes of observations at unprecedented resolution and depth. The challenge is to compare data to simulations so as to test theories and identify those that best match the observations. The data are numerous and rich but are subject to noise and various systematic biases. The simulations are large and detailed but are so costly that the effective sample size is small. Moreover, the mapping from data to simulation to the parameters of interest is typically complicated and ill-conditioned. Although these challenges, and the solutions we propose, apply across the sciences, we will ground our efforts by focusing on cosmology. Here, we are truly entering the era of big data in several senses. Ever larger experiments survey larger and larger cosmological volumes, resulting in enormous amounts of pixel data to process and extract the information of cosmological interest. These include large imaging surveys, which collect all the light in broad wavelength ranges over significant fractions of the sky, and spectroscopic surveys, which measure the light as a function of wavelength to map out 3D distances. These experiments are designed to shed light on the big unanswered questions in cosmology. Distilling the petabytes of data into observed quantities for comparison with theory is a daunting task. But these large experiments do not only result in large amounts of data to analyze: if we wish to compare to theories of physics, then we need increasingly extensive predictions for the observed quantities as well, which requires ever-larger cosmological simulations that take substantial -- and at times prohibitive -- computational resources to produce. Our goal is to develop statistical and machine learning methods for using observed and simulated data to advance machine learning with applications cosmology. In particular, we focus our research efforts on the following tasks: The challenge for automated science is that it is computationally impossible and statistically dangerous to consider every possible model in order to find the best one. We will develop Bayesian Optimization based active-learning methods that accelerate both the execution of the simulations and the search for best-fitting parameters. The key idea is to make the simulations adaptive -- across resolutions, time, and parameters -- using the data to search as the simulation runs; Most ML methods operate on simple finite dimensional feature vectors. However, many cosmology and other science applications require ML methods that can operate on more complex objects as inputs or outputs such as functions, distributions, or sets and point clouds. Our goal is to develop efficient ML methods for this problem and demonstrate their applicability in Cosmology, Astrophysics, and other science problems.

79 ASTRONOMY AND ASTROPHYSICS↗

UNIFI 1-MW Multi-Vendor Grid-Forming Inverter Test Updates

This presentation gives an update on UNIFI 1MW multi-vendor GFM inverter testing. The 1MW demo enters the last stage of the test, which is the integration test. And the main objectives of the integration tests include: 1) explore the interoperability of dispatching a future grid with energy mix (GFM inverters, GFL inverters and synchronous machine). The presentation highlights one example test, ROCOF, and summarizes all the key learning.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Assessment of NDE Technologies for Detection and Characterization of Stress Corrosion Cracking in LWRs

Stress corrosion cracking (SCC) in light water reactors (LWRs) has been a persistent form of degradation in the nuclear industry. Examples of SCC can be found for a range of materials in boiling and pressurized water reactor environments, including carbon steels, stainless steels, and nickel-base stainless alloys. The evolution of SCC is often characterized by a long initiation stage followed by a phase of more rapid crack growth to failure. This provides a relatively short window of opportunity to detect the start of observable SCC, and it is conceivable that SCC could progress from initiation to failure between subsequent examinations when managed by applying periodic in-service inspection techniques. Implementation of advanced aging management paradigms in the current fleet of LWRs will require adaptation of existing measurement technologies and development of new technologies to perform on-line measurements during reactor operation to ensure timely detection of material degradation and to support the implementation of advanced diagnostics and prognostics. This paper considers several non-destructive examination (NDE) technologies with known sensitivity to detection of indicators for SCC initiation and/or propagation, and assesses these technologies with respect to their ability to detect and accurately characterize the significance of an SCC flaw. Potential strategies to improve SCC inspection or monitoring performance are offered to benefit management of SCC degradation in LWRs.

NDE↗

Resilience Metrics and Framework for Distributed Wind Presentation

This presentation communicates information about the MIRACL project Resilience Metrics report and Resilience Framework report. It was created for the 2021 MIRACL advisory board meeting. We propose a three-tiered approach for the resilience framework. At the top level, we consider the time horizons on which resilience will be evaluated and executed. At the middle level, we consider the core functions of resilience, which span across the time horizons. At the lower level, we consider the process steps that correspond to implementing practices for resilience in each of the core functions. The framework considers three time horizons in order to enable organizations to assess and improve their system’s resilience throughout its lifecycle. We call these time horizons the planning, operational, and future stages. The planning stage uses organizational needs and current system evaluation to prepare for potential hazards. The operational stage seeks to execute responses to hazards as prudently and efficiently as possible to maintain system resilience. The future stage seeks to improve on current system resilience and feeds back into the planning stage to promote continuous improvement. While all three time horizons are important when considering a specific topic, the planning and evaluation phase (i.e., what is done in advance of the event) is critical in defining a system’s resilience characteristics and in outlining how a system responds to an event. This framework intentionally emphasizes the planning stage to highlight the overarching emphasis of this effort, not to imply that the other two time-related horizons (i.e., operational and change the future) are less important. The core functions in the framework are identify, prepare, detect, adapt, and recover. These five functions stem from a rigorous analysis of definitions used across the industry, and they represent the core capabilities that an organization must have to enable lifecycle resilience. Within each core function, process steps are described that help walk an organization through the information gathering, evaluation, decision-making, and implementation processes they will need to ensure their resilience goals are maintained throughout the system and the system lifecycle. Also highlighted in the figure is the concept that a resilience framework should be cyclical in nature. Because a system’s resilience is based on finite resources and time, it must continually evolve through this framework’s risk management and capital investment steps at an appropriate level of scope and pace.

17 WIND ENERGY↗

Bifurcations, chaos and adaptive backstepping sliding mode control of a power system with excitation limitation

The bifurcation and Lyapunov exponent for a single-machine-infinite bus system with excitation model are carried out by varying the mechanical power, generator damping factor and the exciter gain, from which periodic motions, chaos and the divergence of system are observed respectively. From given parameters and different initial conditions, the coexisting motions are developed in power system. The dynamic behaviors in power system may switch freely between the coexisting motions, which will bring huge security menace to protection operation. Especially, the angle divergences due to the break of stable chaotic oscillation are found which causes the instability of power system. Finally, a new adaptive backstepping sliding mode controller is designed which aims to eliminate the angle divergences and make the power system run in stable orbits. Numerical simulations are illustrated to verify the effectivity of the proposed method.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Adaptable Data Driven Model Predictive Control for Heat Pipe Microreactors

To establish a technical basis for self-regulating microreactors, a model predictive control (MPC) system is investigated to proactively respond to anomalies and disturbances in anticipation of potential deviations from operating setpoints. Due to the difficulty of developing a physics-based surrogate model that can accurately match plant data in various operating conditions, machine learning algorithms are used in MPC, which allow for learning from both simulation and operation data, thus efficiently describing the targeted transient with arbitrary accuracy. However, one of the biggest concerns in applying ML algorithms like artificial neural networks (ANNs) is that the predictive capabilities of ANN are limited by training data. If there are gaps between the training and target domain, the accuracy of an ANN can degrade significantly when it is used to predict unseen data. To improve the predictive capability of ANN and enable a confident use of data-driven MPCs outside the training data, this study proposes an adaptive data-driven MPC framework. The system will monitor the discrepancy between plant responses and surrogate predictions, fine-tune the ANN-based surrogate when a large discrepancy is detected, and continue MPC operation with updated surrogates. The framework is demonstrated on a point kinetic model for microreactors. The hyperparameters of the update strategy, including layers to update, error thresholds, learning rate discount, and number of data points used for fine-tuning, are optimized so the simulated microreactor is able to follow changes in setpoint with the smallest of deviations.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Safety-Related Instrumentation & Control Pilot Upgrade Initiation Phase Implementation Report

This research report (1) describes the process followed and products developed during the SR I&C Pilot Project Initial Scoping Phase, and (2) captures lessons learned. Exelon Generation and LWRS collaborated to develop a Digital Transformation Strategy as part of a larger Advanced Concept of Operations. The proposed SR I&C Pilot Upgrade provides a foundation stone for this Digital Transformation that will improve plant safety, reliability, and operational performance while lowering plant Total Cost of Ownership (TCO). Initial Scoping Phase activities for this Pilot Project have been performed in accordance with industry processes that have been adapted to better support digital upgrades. These processes include IP-ENG-001, Standard Design Process (SDP) [Reference 2], NISP-EN-04, Standard Digital Engineering Process (SDEP) [Reference 3], and Electric Power Research Institute (EPRI) Report 3002011816, Digital Engineering Guide (DEG). Completing Initial Scoping Phase Engineering and Operations, Licensing, and Project Management Activities was necessary to sufficiently bound the scope, schedule, and estimated cost of the Project to enable utility management to authorize moving into the Conceptual Design Phase. A significant finding of the Business Case Analysis (BCA) methodology developed and applied as part of this effort was that the growth rate of material costs for sustaining the operation of obsolete SR I&C equipment is accelerating. This directly contributed to the Project Economic Analysis created to justify continuing the Project. Project Initial Scoping Phase lessons learned have also been captured to assist the larger industry in understanding the Digital Transformation Strategy and SR I&C Pilot Project Initial Scoping Phase efforts. This is in keeping with the public/private partnership that has been established between the Department of Energy (DOE) and Exelon for this effort with engagement from the NRC. By addressing first-of-a-kind (FOAK) risks and capturing lessons learned, the SR I&C Pilot Upgrade Project addresses technical, regulatory, and business risks to enable subsequent implementers of similar upgrades.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Low-Cost Parallel Algorithms for 2:1 Octree Balance

The logical structure of a forest of octrees can be used to create scalable algorithms for parallel adaptive mesh refinement (AMR), which has recently been demonstrated for several petascale applications. Among various frequently used octree-based mesh operations, including refinement, coarsening, partitioning, and enumerating nodes, ensuring a 2:1 size balance between neighboring elements has historically been the most expensive in terms of CPU time and communication volume. The 2:1 balance operation is thus a primary target to optimize. One important component of a parallel balance algorithm is the ability to determine whether any two given octants have a consistent distance/size relation. Based on new logical concepts we propose fast algorithms for making this decision for all types of 2:1 balance conditions in 2D and 3D. Since we are able to achieve this without constructing any parent nodes in the tree that would otherwise need to be sorted and communicated, we can significantly reduce the required memory and communication volume. In addition, we propose a lightweight collective algorithm for reversing the asymmetric communication pattern induced by non-local octant interactions. We have implemented our improvements as part of the opensource “p4est” software. Benchmarking this code with both synthetic and simulation-driven adapted meshes we are able to demonstrate much reduced runtime and excellent weak and strong scalability. On our largest benchmark problem with 5.13 × 10 11 octants the new 2:1 balance algorithm executes in less than 8 seconds on 112,128 CPU cores of the Jaguar Cray XT5 supercomputer.

Computer Science↗

New adaptive optics control strategy for petawatt-class laser chains

A new generation of ultra-high intensity femtosecond petawatt- and above-class lasers requires new approaches to wavefront corrections. New challenges for adaptive optics consist in overcoming the constraints of potentially bigger diameters, larger amplitude aberrations, faster optics, higher risk of damaging optical components and faster and easier maintenance. Here we present a new technology of a mechanical deformable mirror, which has a large stroke, high temporal stability, low hysteresis, no print-through effect, easy, safe and fast maintenance and an operating frequency up to 10 Hz. We propose the full correction of the final focal spot in the target chamber by a combination of a standard adaptive optics system, a simple focal plane camera and a phase retrieval correction process. We test the reliability of the correction system in terms of intensity variation and wavefront stability. We further verify correction robustness of the method on a large spectral bandwidth and finally perform a focal spot correction on a terawatt laser system in both low and high-power regimes. (paper)

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗