SEARCH · Search NASA

Results for “accelerator performance”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 631 records · Page 35

All sources NASA NTRS DOE OSTI

NASA NTRSDate not supplied

QuAIL Tools for Benchmarking, Analysis and Quantum Algorithm Development

HybridQ and PySA are open-source tools developed by NASA to support benchmarking, analysis and quantum algorithm development in areas such as simulation, optimization and machine learning. These tools leverage classical hardware acceleration via high-performance computing CPU and GPU architectures and support high-performance computing. HybridQ is a highly extensible platform designed to provide a common framework to integrate multiple state-of-the-art techniques to simulate large scale quantum circuits. PySA is an extensible platform to optimize a classical cost function. We provide an outline of each of these open-source tools and highlight projects using each of these tools in contexts of simulation, optimization and machine learning.

Quantum Computing↗

DOE OSTI2020

Real-time Intrusion Detection for High-bandwidth Research Networks using Unsupervised Deep Learning

Nowadays, data generated by large-scale scientific experiments are on the scale of petabytes per month. These data are transferred through dedicated high-bandwidth networks across distributed collaboration sites for processing, storage, and analysis. Like general purpose networks, research networks experience attacks. However, monitoring traffic in such high-bandwidth network environments is challenging with conventional deep packet inspection (DPI)- based network intrusion detection systems (NIDS). Recently, machine learning (ML) algorithms have shown potential in detecting anomalous patterns with better generalizability and faster speed. Compared to the general purpose networks, traffic movement in research network is less volatile and more likely to benefit from the ML-based anomaly analysis. We present ThunderSecure, an efficient AI-powered system for detecting anomalies in 100G research networks. ThunderSecure implements a high-throughput packet processing infrastructure using multi-cores and GPUs. It takes time-series statistics from a variety of network measurements and uses 2D Convolutional Neural Network (CNN) to explore both spatial and temporal dependencies in network data stream. Patterns of science data traffic are learned with a one-class neural network, which blends an Adversarial Autoencoder (AAE) with Gaussian mixture density estimation. Testing traffic flows exhibiting significant deviations from the learned baseline of normality are marked as anomalies. We trained ThunderSecure on hundreds of billions of science data packets recorded from a 100G research network at Fermi National Accelerator Laboratory. The detection performance was evaluated on traffic captured from the same research network days and weeks after the training with different types of attack flows injected. To the best of our knowledge, this is the first ML work in the area of network anomaly detection that has been validated on such extreme scale datasets. Results show that ThunderSecure can recognize science data traffic that are captured long after the training and make nearly certain detection on those with anomalous flows injected, even in the case when the anomaly-to-normal mixing ratio is 0.1%.

Gong, Qian↗

DOE OSTI2023

Electropolishing Study on Nitrogen-doped Niobium Surface

The nitrogen doping (N-doping) process is applied to niobium (Nb) superconducting cavities to enhance their quality factors. The N-doping is followed by an electropolishing process that provides the final surface of the cavities. A controlled EP process is necessary to get the benefit of N-doping and achieve a high accelerating gradient. We have performed electropolishing of N-doped Nb surface under various conditions to understand their impact on the surface. A modified EP process was developed to obtain a smooth pit-free surface.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2025

Artificial Intelligence and Digital Engineering as Enablers for System Engineering in the Energy Sector

Systems engineering is of utmost importance for the success of high-cost, high-complexity megaprojects, which are common in the energy sector. However, the traditional document-centric systems engineering approach tends to be labor-intensive and time-consuming, which has inhibited its full adoption despite proven metrics on its return on investment. However, with the modern approach of digital engineering and technological advancements in artificial intelligence (AI) technologies, the barriers to systems engineering adoption can finally be broken. This paper goes through the systems engineering V-model for lifecycle management and assesses the current state of implementation of digital engineering (especially, mod-el-based systems engineering, digital twins, and digital threads) and AI for each step. It was observed that a combination of digital engineering and AI is being used across different industries to accelerate and optimize systems engineering processes such as concept development, requirements management, architecture definition, system development, verification and validation, operations, and maintenance. Specifically in the energy sector, AI-augmented digital engineering has shown initial potential in accelerated development and deployment, performance optimization, anomaly detection, predictive maintenance, and configuration management. However, challenges remain in integrating DE and AI into an end-to-end system lifecycle management ecosystem safely and reliably. Addressing these challenges and continuously developing impactful tools will enable fast, efficient, and high-frequency deployment of power generation capabilities to keep up with the world’s energy demands and build energy security.

42 - ENGINEERING↗

DOE OSTI2025

Artificial Intelligence and Digital Engineering as Enablers for Systems Engineering in the Energy Sector

42 - ENGINEERING↗

DOE OSTI2021

Computational Offload with BlueField Smart NICs

The recent introduction of a new generation of "smart NICs" have provided new accelerator platforms that include CPU cores or reconfigurable fabric in addition to traditional networking hardware and packet offloading capabilities. While there are currently several proposals for using these smartNICs for low-latency, in-line packet processing operations, there remains a gap in knowledge as to how they might be used as computational accelerators for traditional high-performance applications. This work aims to look at benchmarks and mini-applications to evaluate possible benefits of using a smartNIC as a compute accelerator for HPC applications. We investigate NVIDIA's current-generation BlueField-2 card, which includes eight Arm CPUs along with a small amount of storage, and we test the networking and data movement performance of these cards compared to a standard Intel server host. We then detail how two different applications, YASK and miniMD can be modified to make more efficient use of the BlueField-2 device with a focus on overlapping computation and communication for operations like neighbor building and halo exchanges. Our results show that while the overall compute performance of these devices is limited, using them with a modified miniMD algorithm allows for potential speedups of 5 to 20% over the host CPU baseline with no loss in simulation accuracy.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS2016

Suited and Unsuited Hybrid III Impact Testing and Finite Element Model Characterization

NASA spacecraft design requirements for occupant protection are a combination of the Brinkley Dynamic Response Criteria and injury assessment reference values (IARV) extracted from anthropomorphic test devices (ATD). For the ATD IARVs, the requirements specify the use of the 5th percentile female Hybrid III and the 95th percentile male Hybrid III. Each of these ATDs is required to be fitted with an articulating pelvis (also known as the aerospace pelvis) and a straight spine. The articulating pelvis is necessary for the ATD to fit into spacecraft seats, while the straight spine is required as injury metrics for vertical accelerations are better defined for this configuration. Sled testing of the Hybrid III 5th Percentile Female Anthropomorphic Test Device (ATD) was performed at Wright-Patterson Air Force Base (WAPFB). Two 5th Percentile ATDs were tested, the Air Force Research Lab (AFRL) and NASA owned Hybrid III ATDs with aerospace pelvises. Testing was also conducted with a NASA-owned 95th Percentile Male Hybrid III with aerospace pelvis at WPAFB. Testing was performed using an Orion seat prototype provided by Johnson Space Center (JSC). A 5-point harness comprised of 2 inch webbing was also provided by JSC. For suited runs, a small and extra-large Advanced Crew Escape System (ACES) suit and helmet were also provided by JSC. Impact vectors were combined frontal/spinal and rear/lateral. Some pure spinal and rear axis testing was also performed for model validation. Peak accelerations ranged between 15 and 20-g. This range was targeted because the ATD responses fell close to the IARV defined in the Human-Systems Integration Requirements (HSIR) document. Rise times varied between 70 and 110 ms to assess differences in ATD responses and model correlation for different impact energies. The purpose of the test series was to evaluate the Hybrid III ATD models in Orion-specific landing orientations both with and without a spacesuit. The results of these tests were used by the NASA Engineering and Safety Center (NESC) to validate the finite element model (FEM) of the Hybrid III 5th Percentile Female ATD. Physical test data was compared to analytical predictions from simulations, and modelling uncertainty factors have been determined for each injury metric. Additionally, the test data has been used to further improve the FEM, particularly in the areas of the ATD preload, harness, and suit and helmet effects.

Lawrence, C.↗

NASA NTRS1976

Optimum design considerations of a gust alleviator for aircraft

A gust alleviation system for aircraft flying in turbulent air was analyzed. A vane sensor (with noise) was used to measure vertical gusts, and elevators and flaps were used to reduce the root-mean-square value of the normal accelerations associated with the aircraft response to gusts. Since turbulence has stochastic properties, stochastic control theory was used in the analysis. A quadratic performance-index function involving normal acceleration and control deflections was minimized. Application of the analysis was illustrated by a short take-off and landing (STOL) airplane in flight through turbulent air. Effects of varying the noise characteristics of the vane sensor and of a weighting matrix in the performance-index function were determined. Calculations were performed as required by stochastic control theory to obtain the root-mean-square response of the airplane to turbulence. Results show that good alleviation was calculated when the intensity of the measurement noise was about 3.6 percent of the vane deflection angles.

Oehman, W. I.↗

NASA NTRS2018

Detonation to Deflagration-Mode Transitions in Pulsed Plasma Accelerators

PULSED plasma accelerators typically operate by storing energy in a capacitor bank and then discharging this energy through a gas. The current in such discharges will ionize the gas and produce a strong magnetic field, which interacts with the flowing current to accelerate the plasma through the Lorentz body force. For the present work, two plasma accelerator types employing this general scheme are of interest: the gas-fed pulsed plasma thruster (PPT) 1 and the quasi-steady magnetoplasmadynamic (MPD) accelerator. 2 The gas-fed pulsed plasma accelerator is generally understood as a completely transient device discharging in ∼1-10 μs. When the capacitor bank is discharged through the gas, a current sheet forms at the breech of the thruster and propagates forward under a j × B body force, entraining and accelerating propellant it encounters. This process is sometimes referred to in literature as 'snowplowing' the propellant or accelerating the gas in a detonation-mode because the current sheet representation approximates that of a strong detonation shockwave propagating through the gas. For these devices, acceleration of the initial current sheet ceases when either the current sheet reaches the end of the device and is ejected or when the current in the circuit reverses, striking a new 'crowbar' discharge at the breech and depriving the initial sheet of additional acceleration. In general, PPTs typically claim thrust efficiencies (ratio of jet kinetic energy to input electrical energy) registering in the teens or lower. 3 In the quasi-steady MPD accelerator, the pulse is lengthened to ∼1 ms or longer and maintained at an approximately constant level during discharge through the use of a pulse-forming network (PFN) of capacitors. After an initial transient discharge, which is typically short relative to the overall discharge period, the plasma assumes a relatively steady-state configuration, known as 'quasi-steady' MPD operation. 2 In this state, ionized gas flows through a stationary current channel in a manner that is sometimes referred to as deflagration-mode operation owing to the similarities to deflagration waves in gases. The plasma experiences electromagnetic acceleration as the plasma flows through the current channel towards the exit of the device. Quasi-steady MPD thrusters claim efficiencies up to 50% for certain propellants.4 There has been significant and sustained research over several decades on both gas-fed PPTs and quasi-steady MPD thrusters, however there have been pulsed thrusters that do not appear to exactly fit either classification, instead possessing a mixture of operational qualities characteristic of both thruster variants. The Coaxial High ENerGy (CHENG) thruster by Cheng, et al.5 operated on the short 10 μs timescales characteristic of PPTs, but claimed the high thrust densities, high efficiencies, and low electrode erosion rates that are more consistent with the MPD/deflagration mode of plasma acceleration. Gas-fed PPT research by Ziemer, et al. 3, 6 identified two separate regimes of performance in those thrusters. The regime at higher mass bits (termed Mode I in that work) possessed relatively constant thrust efficiency as a function of mass bit, while the second regime at very low mass bits (termed Mode II) exhibited an increase in efficiency with decreasing mass bit. Work by Poehlmann et al.7 and by Sitaraman and Raja8 sought to understand the performance of the CHENG thruster and the Mode I/Mode II performance in PPTs by modeling the acceleration using the Hugoniot Relation, with the detonation and deflagration modes of plasma acceleration representing two distinct sets of solutions to the relevant conservation laws. In these works, it was proposed that the values of the various controllable parameters determined whether the accelerator would operate in detonation or deflagration mode. Our hypothesized view of the acceleration process in the CHENG thruster and in PPTs experiencing a transition from Mode I to Mode II is inspired by observations of the transition from the PPT mode of operation to the quasi-steady MPD mode. Specifically, the quasi-steady MPD was discovered by driving a PPT to extended pulse lengths. Above a certain pulse length threshold the transient plasma current sheet transitions into a stable plasma acceleration mode that 'replicates in every observable detail steady flow self-field magnetoplasmadynamic acceleration.' 9 In the present work, instead of treating the accelerator as if it were only operating in a single mode during a pulse, we consider the initial stage of the discharge in all cases as a current sheet forming at the breach of the accelerator and moving towards the exit as a detonation wave. If the current sheet reaches the exit of the accelerator before the discharge is completed, the view of the acceleration mode transitions to the deflagration mode-type found in quasi-steady MPD thrusters. In previous work10 we presented a modeling framework that first captured the time-evolution of the current sheet (detonation) mode of the thruster and then transitioned into the quasi-steady MPD (deflagration) mode of plasma acceleration. In the present work, variations of the controllable parameters - specifically the pulsed circuit properties, the amount of mass injected into the thruster, and the relative timing between the initial gas injection and the initiation of the plasma current sheet - will be used to explore the thruster performance. A range of parameters are explored to demonstrate that standard gas-fed pulsed plasma accelerators, the CHENG thruster, and the quasi-steady MPD accelerator are variations of the same device, with the overall acceleration of the plasma depending upon the behavior of the plasma discharge during initial transient phase and the relative lengths of the detonation and deflagration modes of operation.

Polzin, Kurt A.↗

DOE OSTI2022

9.3 Microns: Toward a Next-Generation CO2 Laser for Particle Accelerators

We present realistic theoretical predictions of the performance of the next-generation long-wave infrared (LWIR) laser for advanced particle acceleration research at the Accelerator Test Facility (ATF) of Brookhaven National Laboratory. Two upgrades are planned for the present ATF CO2 laser, which currently produces up to 5 TW peak power in 2 ps pulses at 9.2 μm. For the first upgrade, the deployment of a ten-millijoule solid-state seed laser at 9.3 μm will reduce the pulse duration 4 times, to 500 fs, and increase the peak power to 15 TW. The second upgrade will be the implementation of the postcompression of this pulse. This will reduce the pulse duration to three optical cycles (100 fs) and increase the usable peak power to 25 TW.

43 PARTICLE ACCELERATORS↗

NASA NTRS1992

Physical and digital simulations for IVA robotics

Space based materials processing experiments can be enhanced through the use of IVA robotic systems. A program to determine requirements for the implementation of robotic systems in a microgravity environment and to develop some preliminary concepts for acceleration control of small, lightweight arms has been initiated with the development of physical and digital simulation capabilities. The physical simulation facilities incorporate a robotic workcell containing a Zymark Zymate II robot instrumented for acceleration measurements, which is able to perform materials transfer functions while flying on NASA's KC-135 aircraft during parabolic manuevers to simulate reduced gravity. Measurements of accelerations occurring during the reduced gravity periods will be used to characterize impacts of robotic accelerations in a microgravity environment in space. Digital simulations are being performed with TREETOPS, a NASA developed software package which is used for the dynamic analysis of systems with a tree topology. Extensive use of both simulation tools will enable the design of robotic systems with enhanced acceleration control for use in the space manufacturing environment.

Hinman, Elaine↗

NASA NTRS1989

Electrostatic Plasma Accelerator (EPA)

The Electrostatic Plasma Accelerator (EPA) is a thruster concept which promises specific impulse levels between low power arcjets and those of the ion engine while retaining the relative simplicity of the arcjet. The EPA thruster produces thrust through the electrostatic acceleration of a moderately dense plasma. No accelerating electrodes are used and the specific impulse is a direct function of the applied discharge voltage and the propellant atomic mass. The goal of the present program is to demonstrate feasibility of the EPA thruster concept through experimental and theoretical investigations of the EPA acceleration mechanism and discharge chamber performance. Experimental investigations will include operating the test bed ion (TBI) engine as an EPA thruster and parametrically varying the thruster geometry and operating conditions to quantify the electrostatic plasma acceleration effect. The theoretical investigations will include the development of a discharge chamber model which describes the relationships between the engine size, plasma properties, and overall performance. For the EPA thruster to be a viable propulsion concept, overall thruster efficiencies approaching 30% with specific impulses approaching 1000 s must be achieved.

Brophy, John R.↗

DOE OSTI2022

GronOR: Scalable and Accelerated Nonorthogonal Configuration Interaction for Molecular Fragment Wave Functions

GronOR is a program package for nonorthogonal configuration interaction calculations. Electronic wave functions are constructed in terms of antisymmetrized products of multiconfiguration molecular fragment wave functions. The computational complexity of the nonorthogonal methodologies implemented in GronOR applied to large molecular assemblies requires a design that takes full advantage of massively parallel supercomputer architectures and accelerator technologies. This work describes the implementation strategy and resulting performance characteristics. In addition to parallelization and acceleration, the software development strategy includes aspects of fault resiliency and heterogeneous computing. The program was designed for large-scale supercomputers but also runs effectively on small clusters and workstations for small molecular systems. GronOR is available as open source to the scientific community.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2025

Virtual to Physical: Reinforcement Learning to Optimize SNS Particle Accelerator Controls

Complex accelerators must have control systems that can handle dynamic nonlinear environments. This makes traditional control methods unsuitable as they can struggle to adapt to these uncertainties. This provides an ideal environment for reinforcement learning algorithms as they are adaptable and generalizable. We present a reinforcement learning pipeline that can effectively handle the dynamics of a complex accelerator. We test and prove our pipelines capabilities on multiple environments including the Spallation Neutron Source (SNS) and the Beam Test Facility (BTF) at Oakridge National Lab (ORNL). Due to the limited time available to train an online algorithm like reinforcement learning on a real accelerator, we utilize a virtual twin accelerator (VIRAC) developed by ORNL to pretrain the policy and show its ability to converge in the virtual environment. We then test the adaptability of the pretrained RL model by applying it on the real accelerator and comparing the results. Utilizing our Scientific Optimization and Controls Toolkit (SOCT) and open-source standards such as Gymnasium we create and solve for a MEBT orbit correction problem in the SNS and an emittance maximization problem in the BTF. We show how Twin Delayed Deep Deterministic Policy Gradient (TD3) can solve this optimization environment in the virtual accelerator and transfer this policy onto the real accelerator for inference and model retraining. We show how reinforcement learning can be utilized as a control system for complex accelerators and provide a model pipeline for how an implementation performs and can be adapted to new accelerator control problems.

Kasparian, Armen [Thomas Jefferson National Accele↗

NASA NTRS1993

Acceleration Studies

Work to support the NASA MSFC Acceleration Characterization and Analysis Project (ACAP) was performed. Four tasks (analysis development, analysis research, analysis documentation, and acceleration analysis) were addressed by parallel projects. Work concentrated on preparation for and implementation of near real-time SAMS data analysis during the USMP-1 mission. User support documents and case specific software documentation and tutorials were developed. Information and results were presented to microgravity users. ACAP computer facilities need to be fully implemented and networked, data resources must be cataloged and accessible, future microgravity missions must be coordinated, and continued Orbiter characterization is necessary.

Rogers, Melissa J. B.↗

DOE OSTI2022

Analysis of cell-based diffusion acceleration for the slice balance approach

In this work, we perform analysis on the use of cell-based diffusion acceleration methodologies to accelerate the convergence of transport solutions discretized with the slice balance approach (SBA) on unstructured polygonal grids.We investigated both linear diffusion synthetic acceleration (DSA) and non linear diffusion acceleration (NDA), including its partial-current variant (pNDA). DSA and NDA were both shown to diverge for intermediate ranges of mesh optical thicknesses. However, pNDA and Krylov methods like GMRES and Broyden stabilized the acceleration schemes, including problems with degenerate cells formed by mesh refinement. (author)

42 ENGINEERING↗

NASA NTRS1990

Experiments in cooperative-arm object manipulation with a two-armed free-flying robot

Developing computed-torque controllers for complex manipulator systems using current techniques and tools is difficult because they address the issues pertinent to simulation, as opposed to control. A new formulation of computed-torque (CT) control that leads to an automated computer-torque robot controller program is presented. This automated tool is used for simulations and experimental demonstrations of endpoint and object control from a free-flying robot. A new computed-torque formulation states the multibody control problem in an elegant, homogeneous, and practical form. A recursive dynamics algorithm is presented that numerically evaluates kinematics and dynamics terms for multibody systems given a topological description. Manipulators may be free-flying, and may have closed-chain constraints. With the exception of object squeeze-force control, the algorithm does not deal with actuator redundancy. The algorithm is used to implement an automated 2D computed-torque dynamics and control package that allows joint, endpoint, orientation, momentum, and object squeeze-force control. This package obviates the need for hand-derivation of kinematics and dynamics, and is used for both simulation and experimental control. Endpoint control experiments are performed on a laboratory robot that has two arms to manipulate payloads, and uses an air bearing to achieve very-low drag characteristics. Simulations and experimental data for endpoint and object controllers are presented for the experimental robot - a complex dynamic system. There is a certain rather wide set of conditions under which CT endpoint controllers can neglect robot base accelerations (but not motions) and achieve comparable performance including base accelerations in the model. The regime over which this simplification holds is explored by simulation and experiment.

Koningstein, Ross↗

DOE OSTI2024

IRIS-GNN: Leveraging Graph Neural Networks for Scheduling on Truly Heterogeneous Runtime Systems

The diversity of accelerators in computer systems poses significant challenges for software developers, such as managing vendor-specific compiler toolchains, code fragmentation requiring different kernel implementations, and performance portability issues. To address these, the Intelligent Runtime System (IRIS) was developed. IRIS works across various systems, from smartphones to supercomputers, enabling automatic performance scaling based on available accelerators. It introduces abstract tasks for seamless execution transitions between accelerators while ensuring memory consistency and task dependencies. Although IRIS simplifies system details, optimal dynamic scheduling still requires user input to understand workload structures. To address this, we introduce a new scheduling policy for IRIS, termed IRIS-GNN, which is the first IRIS hybrid policy that operates in conjunction with the dynamic policies. This policy employs a Graph-Neural Network (GNN) to conduct Graph Classification of any task graphs submitted to IRIS. This GNN analyzes the structure and attributes of the task graph, categorizing it as either locality, concurrency, or mixed. This classification subsequently guides the selection of the dynamic policy used by IRIS. We provide a comparison of the performance of IRIS-GNN against the complete spectrum of IRIS’s dynamic policies, assess the overhead introduced by the GNN within this scheduling framework, and ultimately explore its practical application in real-world scenarios.

Johnston, Beau↗