Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “hardware efficiency”

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 91 records · Page 5
All sourcesNASA NTRSDOE OSTI
NASA NTRS2005

A Plug and Play GNC Architecture Using FPGA Components

The goal of Plug and Play, or PnP, is to allow hardware and software components to work together automatically, without requiring manual setup procedures. As a result, new or replacement hardware can be plugged into a system and automatically configured with the appropriate resource assignments. However, in many cases it may not be practical or even feasible to physically replace hardware components. One method for handling these types of situations is through the incorporation of reconfigurable hardware such as Field Programmable Gate Arrays, or FPGAs. This paper describes a phased approach to developing a Guidance, Navigation, and Control (GNC) architecture that expands on the traditional concepts of PnP, in order to accommodate hardware reconfiguration without requiring detailed knowledge of the hardware. This is achieved by establishing a functional based interface that defines how the hardware will operate, and allow the hardware to reconfigure itself. The resulting system combines the flexibility of manipulating software components with the speed and efficiency of hardware.

KrishnaKumar, K.↗
DOE OSTI2022

ExaSGD: 2022 Kernel Thrust Activities

The Kernel Thrust milestone ADSE22-407 covers the development of device-capable optimization algorithms and solvers technologies required by the ExaSGD project’s software stack in order to solve security-constrained alternating current optimal power flow (SC-ACOPF) problems on emerging exascale architectures. To this extent, in FY22 the main objective of the Kernel Thrust was (i) provide sparse optimization solver that runs efficiently on hardware accelerator devices (i.e., NVIDIA and AMD GPUs) to perform intra-node computations, (ii) strengthen the reliability and increase the performance of the mixed-dense sparse (MDS) solver of HiOp for deployment on the FY22 target architectures, Summit and Crusher, and (iii) increase performance by improving the mathematical algorithm and refining the parallel MPI-based implementation of the coarse-grain parallel solver HiOp-PriDec for capabilities deployment on the FY22 target architectures, Summit and Crusher. This document presents the developments and contributions done by the Kernels Thrust Team in FY22 toward completion of the above-mentioned objectives. These contributions progressed along four main development (sub)thrusts: (1) Design and implementation of a sparse optimization solver for use on hardware accelerators; (2) Improvement of the mathematical algorithm and of the parallel implementation of HiOp-PriDec to ensure readiness and efficient coarse-grain parallelism for FY23 target exascale machine; and (3) Support Software and Application Development Thrusts of the exaSGD project in their deployment of the project’s software stack on AMD- and NVIDIA-based architectures. The development of the sparse optimization solver (thrust 1 above) was new in FY22 and resulted in a new sparse solver in HiOp (available as of version 0.6). The second development thrust was a continuation of the efforts from FY21 and improved the mathematical algorithm and the communication strategy of the HiOp-PriDec solver. The last developement thrust is a large collaborative effort. Namely, the project’s teams from multiple labs (LLNL, PNNL, ORNL, and NREL) performed large-scale demonstration of the ExaSGD software stack, namely the optimization solvers of HiOp interfaced with the modeling front-end ExaGO and the stochastic sampler PowerScenarios. These demonstration efforts solved large-scale instances of the SC-ACOPF challenge problem of medium network sizes (10, 000-bus system) and large number of contingencies on Summit (NVIDIA accelerators) and Crusher (AMD accelerators) systems at ORNL.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1982

Energy efficient engine. Low pressure turbine test hardware detailed design report

The low pressure turbine for the energy efficient engine is a five-stage configuration with moderate aerodynamic loading incorporating advanced features of decambered airfoils and extended blade overlaps at platforms and shrouds. Mechanical integrity of 18,000 hours on flowpath components and 36,000 hours on all other components is achieved along with no aeromechanical instabilities within the steady-state operating range. Selection of a large number (156) of stage 4 blades, together with an increased stage 4 vane-to-blade gap, assists in achieving FAR 36 acoustic goals. Active clearance control (ACC) of gaps at blade tips and interstage seals is achieved by fan air cooling judiciously applied at responsive locations on the casing. This ACC system is a major improvement in preventing deterioration of the 0.0381 cm (0.015 in.) clearances required to meet the integrated-core/low-spool turbine efficiency goal of 91.1% and the light propulsion system efficiency goal of 91.7%.

Cherry, D. G.↗
NASA NTRS1982

Energy efficient engine high pressure turbine test hardware detailed design report

The high pressure turbine configuration for the Energy Efficient Engine is built around a two-stage design system. Moderate aerodynamic loading for both stages is used to achieve the high level of turbine efficiency. Flowpath components are designed for 18,000 hours of life, while the static and rotating structures are designed for 36,000 hours of engine operation. Both stages of turbine blades and vanes are air-cooled incorporating advanced state of the art in cooling technology. Direct solidification (DS) alloys are used for blades and one stage of vanes, and an oxide dispersion system (ODS) alloy is used for the Stage 1 nozzle airfoils. Ceramic shrouds are used as the material composition for the Stage 1 shroud. An active clearance control (ACC) system is used to control the blade tip to shroud clearances for both stages. Fan air is used to impinge on the shroud casing support rings, thereby controlling the growth rate of the shroud. This procedure allows close clearance control while minimizing blade tip to shroud rubs.

Halila, E. E.↗
DOE OSTI2021

A Survey: Handling Irregularities in Neural Network Acceleration with FPGAs

In the last decade, Artificial Intelligence (AI) through Deep Neural Networks (DNNs) has penetrated virtually every aspect of science, technology, and business. Many types of DNNs have been and continue to be developed, including Convolutional Neural Networks (CNNs), Recurrent Neural Net- works (RNNs), and Graph Neural Networks (GNNs). The overall problem for all of these Neural Networks (NNs) is that their target applications generally pose stringent constraints on latency and throughput, while also having strict accuracy requirements. There have been many previous efforts in creating hardware to accelerate NNs. The problem designers face is that optimal NN models typically have significant irregularities, making them hardware-unfriendly. In this paper, we first define the problems in NN acceleration by characterizing common irregularities in NN processing into 4 types; then we summarize the existing works that handle the four types of irregularities efficiently using hardware, especially FPGAs; finally, we provide a new vision of next-generation FPGA-based NN acceleration: that the emerging heterogeneity in the next-generation FPGAs is the key to achieving higher performance.

Geng, Tong↗
NASA NTRS1991

Fiber Optic Tactical Local Network (FOTLAN)

A 100 Mbit/s FDDI (fiber distributed data interface) network interface unit is described that supports real-time data, voice and video. Its high-speed interrupt-driven hardware architecture efficiently manages stream and packet data transfer to the FDDI network. Other enhancements include modular single-mode laser-diode fiber optic links to maximize node spacing, optic bypass switches for increased fault tolerance, and a hardware performance monitor to gather real-time network diagnostics.

Bergman, L. A.↗
NASA NTRS1991

A high performance FDDI NIU for streaming voice, video, and data

A 100-Mbit/s FDDI network interface unit (NIU) is described that supports real-time data, voice and video. Its high-speed interrupt-driven hardware architecture efficiently manages stream and packet data transfers to the FDDI network. Other enhancements include modular single-mode laser-dioce fiber optic links to maximize node spacing, optic bypass switches for increased fault tolerance, and a hardware performance monitor to gather real-time network diagnostics.

Bergman, L. A.↗
DOE OSTI2025

Solving sparse finite element problems on neuromorphic hardware

The finite element method (FEM) is one of the most important and ubiquitous numerical methods for solving partial differential equations (PDEs) on computers for scientific and engineering discovery. Applying the FEM to larger and more detailed scientific models has driven advances in high-performance computing for decades. Here we demonstrate that scalable spiking neuromorphic hardware can directly implement the FEM by constructing a spiking neural network that solves the large, sparse, linear systems of equations at the core of the FEM. We show that for the Poisson equation, a fundamental PDE in science and engineering, our neural circuit achieves meaningful levels of numerical accuracy and close to ideal scaling on modern, inherently parallel and energy-efficient neuromorphic hardware, specifically Intel’s Loihi 2 neuromorphic platform. We illustrate extensions to irregular mesh geometries in both two and three dimensions as well as other PDEs such as linear elasticity. Our spiking neural network is constructed from a recurrent network model of the brain’s motor cortex and, in contrast to black-box deep artificial neural network-based methods for PDEs, directly translates the well-understood and trusted mathematics of the FEM to a natively spiking neuromorphic algorithm.

Applied mathematics↗
DOE OSTI2025

LuGo: An enhanced quantum phase estimation implementation

Quantum Phase Estimation (QPE) is a cardinal algorithm in quantum computing that plays a crucial role in various applications, including cryptography, molecular simulation, and solving systems of linear equations. However, the standard implementation of QPE faces challenges related to time complexity and circuit depth, which limit its practicality for large-scale computations. We introduce LuGo, a novel framework designed to enhance the performance of QPE by reducing circuit duplication, as well as using parallelization techniques to achieve faster generation of the QPE circuit and gate reduction. We validate the effectiveness of our framework by generating quantum linear solver circuits, which require both QPE and inverse QPE, to solve linear systems of equations. LuGo achieves significant improvements in both computational efficiency and hardware requirements without compromising on accuracy. Compared to a standard QPE implementation, LuGo reduces time consumption to generate a circuit that solves a 2 6 × 2 6 system matrix by a factor of 50.68 and over 31× reduction of quantum gates and circuit depth, with no fidelity loss on an ideal quantum simulator. Furthermore, we demonstrated the versatility and scalability of LuGo enabled HHL algorithm by simulating a canonical Hele-Shaw fluid problem using a quantum simulator. With these advantages, LuGo paves the way for more efficient implementations of QPE, enabling broader applications across several quantum computing domains.

Quantum algorithm↗
NASA NTRS2017

Virtualized Multi-Mission Operations Center (vMMOC) and its Cloud Services

His presentation will cover, the current and future, technical and organizational opportunities and challenges with virtualizing a multi-mission operations center. The full deployment of Goddard Space Flight Centers (GSFC) Virtualized Multi-Mission Operations Center (vMMOC) is nearly complete. The Space Science Mission Operations (SSMO) organizations spacecraft ACE, Fermi, LRO, MMS(4), OSIRIS-REx, SDO, SOHO, Swift, and Wind are in the process of being fully migrated to the vMMOC. The benefits of the vMMOC will be the normalization and the standardization of IT services, mission operations, maintenance, and development as well as ancillary services and policies such as collaboration tools, change management systems, and IT Security. The vMMOC will also provide operational efficiencies regarding hardware, IT domain expertise, training, maintenance and support.The presentation will also cover SSMO's secure Situational Awareness Dashboard in an integrated, fleet centric, cloud based web services fashion. Additionally the SSMO Telemetry as a Service (TaaS) will be covered, which allows authorized users and processes to access telemetry for the entire SSMO fleet, and for the entirety of each spacecrafts history. Both services leverage cloud services in a secure FISMA High and FedRamp environment, and also leverage distributed object stores in order to house and provide the telemetry. The services are also in the process of leveraging the cloud computing services elasticity and horizontal scalability. In the design phase is the Navigation as a Service (NaaS) which will provide a standardized, efficient, and normalized service for the fleet's space flight dynamics operations. Additional future services that may be considered are Ground Segment as a Service (GSaaS), Telemetry and Command as a Service (TCaaS), Flight Software Simulation as a Service, etc.

Ido, Haisam Kassim↗
DOE OSTI2023

Efficient human activity recognition with spatio-temporal spiking neural networks

In this study, we explore Human Activity Recognition (HAR), a task that aims to predict individuals' daily activities utilizing time series data obtained from wearable sensors for health-related applications. Although recent research has predominantly employed end-to-end Artificial Neural Networks (ANNs) for feature extraction and classification in HAR, these approaches impose a substantial computational load on wearable devices and exhibit limitations in temporal feature extraction due to their activation functions. To address these challenges, we propose the application of Spiking Neural Networks (SNNs), an architecture inspired by the characteristics of biological neurons, to HAR tasks. SNNs accumulate input activation as presynaptic potential charges and generate a binary spike upon surpassing a predetermined threshold. This unique property facilitates spatio-temporal feature extraction and confers the advantage of low-power computation attributable to binary spikes. We conduct rigorous experiments on three distinct HAR datasets using SNNs, demonstrating that our approach attains competitive or superior performance relative to ANNs, while concurrently reducing energy consumption by up to 94%.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2025

Optical neural engine for solving scientific partial differential equations

Abstract Solving partial differential equations (PDEs) is the cornerstone of scientific research and development. Data-driven machine learning (ML) approaches are emerging to accelerate time-consuming and computation-intensive numerical simulations of PDEs. Although optical systems offer high-throughput and energy-efficient ML hardware, their demonstration for solving PDEs is limited. Here, we present an optical neural engine (ONE) architecture combining diffractive optical neural networks for Fourier space processing and optical crossbar structures for real space processing to solve time-dependent and time-independent PDEs in diverse disciplines, including Darcy flow equation, the magnetostatic Poisson’s equation in demagnetization, the Navier-Stokes equation in incompressible fluid, Maxwell’s equations in nanophotonic metasurfaces, and coupled PDEs in a multiphysics system. We numerically and experimentally demonstrate the capability of the ONE architecture, which not only leverages the advantages of high-performance dual-space processing for outperforming traditional PDE solvers and being comparable with state-of-the-art ML models but also can be implemented using optical computing hardware with unique features of low-energy and highly parallel constant-time processing irrespective of model scales and real-time reconfigurability for tackling multiple tasks with the same architecture. The demonstrated architecture offers a versatile and powerful platform for large-scale scientific and engineering computations.

Tang, Yingheng (ORCID:0009000153622546)↗
DOE OSTI2021

Multi-Objective Hyperparameter Optimization for Spiking Neural Network Neuroevolution

Neuroevolution has had significant success over recent years, but there has been relatively little work applying neuroevolution approaches to spiking neural networks (SNNs). SNNs are a type of neural network that includes temporal processing component, are not easily trained using other methods, and can be deployed into energy-efficient neuromorphic hardware. In this work, we investigate two evolutionary approaches for training SNNs. We explore the impact of the hyperparameters of the evolutionary approaches, including tournament size, population size, and representation type, on the performance of the algorithms. We present a multi-objective Bayesian-based hyperparameter optimization approach to tune the hyperparameters to produce the most accurate and smallest SNNs. We show that the hyperparameters can significantly affect the performance of these algorithms. We also perform sensitivity analysis and demonstrate that every hyperparameter value has the potential to perform well, assuming other hyperparameter values are set correctly.

Parsa, Maryam↗
DOE OSTI2021

ExaSGD: 2021 Kernel Thrust Activities

The Kernel Thrust milestone ADSE22-214 covers the development of device-capable optimization algorithms and solvers technologies required by the ExaSGD project’s software stack in order to solve security-constrained alternating current optimal power flow (SC-ACOPF) problems on emerging exascale architectures. To this extent, in FY21 the main objective of the Kernel Thrust was (i) provide robust optimization solver(s) that run efficiently on hardware accelerator devices (i.e., NVIDIA and AMD GPUs) to perform intra-node computations and (ii) provide coarse-grain parallel optimization capabilities that exploit the decomposition opportunities present in the SC-ACOPF challenge problems to provide exascale-capable solvers.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

AOI [1] Advanced Manufacturing of Ceramic Anchors with Embedded Sensors for Process and Health Monitoring of Coal Boilers

Researchers at West Virginia University (WVU) developed methods to fabricate and test ceramic anchors with an embedded sensor technology for monitoring the health and processing conditions within pulverized coal (PC) and fluidized-bed combustion (FBC) boiler systems. The technology included the development of advanced manufacturing processes for 2D/3D printing electroceramic (conductive ceramic) sensor designs within the ceramic anchor microstructure during the manufacturing process. This advanced manufacturing process would allow for the precise control of local microstructure and composition in order to engineer layer-by-layer any protective and electrically active materials within the refractory anchor. This 3D printing technology would permit the rapid and controlled design of the refractory microstructure and embedded sensor design throughout the volume of the ceramic anchor. The work also included a method to interconnect the sensors to boiler shell through the anchor clamp, where the sensor signals will be processed by low-power electronics and transmitted wirelessly to a central processing hub. The end-goal of the program was to produce a ceramic anchor sensor system which would be ready for implementation within a coal boiler, and/or other similar refractory liner systems (such as that in the glass and metal manufacturing areas). The project objectives were to: 1) Define the chemical and microstructural stability, in addition to the electrical properties, of oxide and non-oxide ceramic composites to be embedded within the ceramic anchor compositions that may operate up to 1400ºC; 2) Develop and implement the 2D/3D printing technology to pattern and control the microstructure of the ceramic anchor and embedded sensor circuits; 3) Develop an interconnect technology which will permit easy installation of the ceramic anchors and signal collection at the boiler shell; 4) Develop low power analog electronics and wireless communication hardware to efficiently collect the sensor signal at each processing unit and transmit data to a central hub for data analysis; 5) Demonstrate the smart ceramic anchor system for temperature and liner fracture within a high-temperature processing unit, such as a boiler furnace or glass melting furnace floor/wall liner.

20 FOSSIL-FUELED POWER PLANTS↗
NASA NTRS1988

Energy Efficient Engine: High-pressure compressor test hardware detailed design report

The objective of the NASA Energy Efficient Engine program is to identify and verify the technology required to achieve significant reductions in fuel consumption and operating cost for future commercial gas turbine engines. The design and analysis is documented of the high pressure compressor which was tested as part of the Pratt and Whitney effort under the Energy Efficient Engine program. This compressor was designed to produce a 14:1 pressure ratio in ten stages with an adiabatic efficiency of 88.2 percent in the flight propulsion system. The corresponding expected efficiency for the compressor component test rig is 86.5 percent. Other performance goals are a surge margin of 20 percent, a corrected flow rate of 35.2 kg/sec (77.5 lb/sec), and a life of 20,000 missions and 30,000 hours. Low loss, highly loaded airfoils are used to increase efficiency while reducing the parts count. Active clearance control and case trenches in abradable strips over the blade tips are included in the compressor component design to further increase the efficiency potential. The test rig incorporates variable geometry stator vanes in all stages to permit maximum flexibility in developing stage-to-stage matching. This provision precluded active clearance control on the rear case of the test rig. Both the component and rig designs meet or exceed design requirements with the exception of life goals, which will be achievable with planned advances in materials technology.

Howe, David C.↗
NASA NTRS1992

Development of a machine vision guidance system for automated assembly of space structures

The topics are presented in viewgraph form and include: automated structural assembly robot vision; machine vision requirements; vision targets and hardware; reflective efficiency; target identification; pose estimation algorithms; triangle constraints; truss node with joint receptacle targets; end-effector mounted camera and light assembly; vision system results from optical bench tests; and future work.

Eric G Cooper↗
NASA NTRS2010

Use of Drying Technologies for Resource Recovery from Solid Wastes and Brines

Long term storage of unprocessed biological wastes and human wastes can present major health issues and a loss of potential resources. Space vehicles and planetary habitats are typically resource-scarce or resource-limited environments for long-term human habitation. To-date, most of the resources will need to be supplied from Earth, but this may not be possible for long duration human exploration. Based on present knowledge, there is only very limited in-situ resources on planetary habitats. Hence, the opportunity to "live off the land" in a planetary habitat is limited. However, if we assume that wastes generated by human explorers are viewed as resources, there is great potential to utilize and recycle them, thereby reducing the requirements for supply Earth and enabling the "live off the land" exploration scenario. Technologies used for the recovery of resources from wastes should be reliable, safe, easy to operate, fail-proof, modular, automated and preferably multifunctional in being capable of handling mixed solid and liquid wastes. For a lunar habitat, energy does not appear to be the major driving factor amongst the technologies studied. Instead, reliability appears to be more important[1] . This paper reports studies to date on drying technologies to remove water from solid wastes and brines. Experimental performance data obtained for recovery water from wastes and brine are presented. Simplicity of operation of hardware and energy efficiency are discussed. Some improvements and modifications to hardware were performed. Hopefully, this information will assist in future efforts in the "downselection" of technologies for recovery of water and resources from solid wastes and brines.

Wignarajah, Kanapathipillai↗