Search NASA⌕ Search

SEARCH · Search NASA

Results for “energy efficient computing”

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 361 records · Page 20

HPE ultralit project (ARPA-E open program 2018) final report 15

The goal of the program was to build a fully integrated optical transceiver with >1 Tb/s and <1.5 pJ/bit operating at 50 ◦ C. Optical transceivers are critical components in high-performance-computers (HPC) and data centers, and the details of their implementation has a big impact on the total power consumption (energy efficiency) of an HPC system. Our proposed transceiver used three key enabling technologies. Firstly, SiGe avalanche photodetectors have record-low sensitivities, meaning that they can reach low bit error rates with very little light input. As a result, we can drive our lasers at a lower drive current, thus saving electrical power. Secondly, we use MOS-based capacitive tuning in our deinterleaver, modulator, and demultiplexer. Capacitive tuning allows for the tuning of photonic elements with zero static power consumption. Thirdly, we use quantum dots as the gain material in our light source. This allows us to efficiently use our light source at temperatures that are typically encountered in an HPCsystem. Our proposed optical transceiver consisted of a quantum dot comb laser as a light source, a booster SOA, MOS-based deinterleavers, MOS based ring modulators, MOS-based ring demultiplexers, and SiGe APDs.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

2024 Second Half Semi Annual Report: Modeling plasticity-mediated flow in metals with pressurized cavities

The objective is to better predict the bulk-scale mechanical behavior of porous metals that have over pressurized cavities (e.g., irradiated metals with helium bubbles) by quantifying the complex coupling among cavity aspects (e.g., size distribution, inhomogeneous overpressure values, spatial arrangement) and metal properties (e.g., rate-dependency, crystallographic lattice). This requires up-scaling local mechanical fields from the single crystal scale and will be accomplished using a homogenization approach that combines full-field numerical simulations, analytical formalisms, and physics-informed machine learning to produce symbolically-defined constitutive equations (e.g., gauge functions). These equations will satisfy the objective because they enable computationally efficient predictions that approach the accuracy of computationally expensive full-field numerical simulations, abide by theoretical requirements (e.g., conservation of energy, work conjugacy), and retain the transparency of analytical models.

36 MATERIALS SCIENCE↗

Initial postbuckling analysis of elastoplastic thin-shear structures

The design of thin shell structures with respect to elastoplastic buckling requires an extended analysis of the influence of initial imperfections. For conservative design, the most critical defect should be assumed with the maximum allowable magnitude. This defect is closely related to the initial postbuckling behavior. An algorithm is given for the quasi-static analysis of the postbuckling behavior of structures that exhibit multiple buckling points. the algorithm based upon an energy criterion allows the computation of the critical perturbation which will be employed for the definition of the critical defect. For computational efficiency, the algorithm uses the reduced basis technique with automatic update of the modal basis. The method is applied to the axisymmetric buckling of cylindrical shells under axial compression, and conclusions are given for future research.

E G Carnoy↗

Simulation of proton-induced energy deposition in integrated circuits

A time-efficient simulation technique was developed for modeling the energy deposition by incident protons in modern integrated circuits. To avoid the excessive computer time required by many proton-effects simulators, a stochastic method was chosen to model the various physical effects responsible for energy deposition by incident protons. Using probability density functions to describe the nuclear reactions responsible for most proton-induced memory upsets, the simulator determines the probability of a proton hit depositing the energy necessary for circuit destabilization. This factor is combined with various circuit parameters to determine the expected error-rate in a given proton environment. An analysis of transient or dose-rate effects is also performed. A comparison to experimental energy-disposition data proves the simulator to be quite accurate for predicting the expected number of events in certain integrated circuits.

Fernald, Kenneth W.↗

Energy Efficient Engine Low Pressure Subsystem Flow Analysis

The objective of this project is to provide the capability to analyze the aerodynamic performance of the complete low pressure subsystem (LPS) of the Energy Efficient Engine (EEE). The analyses were performed using three-dimensional Navier-Stokes numerical models employing advanced clustered processor computing platforms. The analysis evaluates the impact of steady aerodynamic interaction effects between the components of the LPS at design and off-design operating conditions. Mechanical coupling is provided by adjusting the rotational speed of common shaft-mounted components until a power balance is achieved. The Navier-Stokes modeling of the complete low pressure subsystem provides critical knowledge of component aero/mechanical interactions that previously were unknown to the designer until after hardware testing.

Hall, Edward J.↗

Disruptive Technologies and Their Putative Impacts Upon Society and Aerospace- Entering The Virtual Age

Developments in technology over the recent decades have been extraordinary. They include the IT, bio, nano, and now quantum and energetics technology arenas and their many combinatorial interactions and impacts. In the main, these are at the frontiers of the small and in a combinational, synergistic feeding frenzy with each other. They fall under the broad category of Disruptive Technologies and have greatly altered society. The outlook for the runout of these and other technology developments augers mid-term to later alterations in components of the human existence theorem, including the requirement to work for our living and our physiological makeup and longevity (Ref 1). The IT revolution began in the 1950s with the development of solid-state electronics. The biologics revolution began later in the 1960s and 1970s with DNA and genomics, and the nano revolution in the 1990s with self-forming nano systems and carbon nanotubes. Quantum technology is now developing rapidly, aided by enabling nano systems, and the energetics revolution is providing ever more efficient and less expensive renewable energy sources. The IT revolution has produced improvements of an astounding eleven orders of magnitude in computing speed since the late 1950s. As we shift from silicon to biological, optical, nano, molecular, and atomic computing, improvements of some 4 orders of magnitude are evidently possible from either optical or DNA computing [Refs 2and 3], then there are combinatorials. Then there is quantum computing, under development worldwide for an increasing number of applications and proffering phenomenal capabilities. The current fastest computers are considerably beyond human brain speed. Machine intelligence is developing well after decades of inadequate machine capability, now no longer the case, and a detour into expert systems. Researchers in machine intelligence are now pursuing deep learning approaches using neural nets, which are proving to be extremely useful. Some believe the frontier of potential human-level machine intelligence may be found in biomimetics and brain-emulation approaches. There is even a possibility of “emergence”—i.e., when the machine intelligence is complex enough that it “wakes up,” as when human intelligence emerged via evolution during the million-plus years of the hunter-gatherer epoch [ Ref 4]. In fact, some posit that human intelligence can be improved upon and is only a cul-de-sac of what is conceivable. The IT revolution has produced massive changes in human society and economics—from the Internet, enabling the rapid expansion of knowledgeability (and even what is knowable), to an increasingly pervasive trend of “tele-everything.” The extraordinary compilation, storage, and availability of truly massive amounts of information could, when combined with AI and under the mantra of “big data,” greatly improve many of our technical and commercial processes and their content including elucidating new heuristic governing laws.

Dennis M. Bushnell↗

Alfalfa Virtual Building Service: Software Engineering Best Practices Applied to Runtime Interaction with Building Energy Models

Buildings are active participants in increasingly complex energy systems. Building Energy Modeling (BEM) has a key role to play in planning and de-risking an equitable energy transition, with BEM-backed "virtual buildings" critical path for diverse applications that include workforce training tools, Hardware-in-the-Loop (HIL) experimentation to study equipment performance under a range of conditions, Control-Hardware-in-the-Loop (CHIL) experimentation to de-risk commercial control implementations at equipment through grid orchestration levels, and integration of dynamic load profiles into grid modeling tools for energy system experimentation at the urban scale. Modeling requirements vary across these applications, but many software engineering tasks do not. The Alfalfa Virtual Building Service (AVBS, see https://github.com/NREL/alfalfa/wiki) is an open-source web service that solves these common tasks robustly in one place, providing a foundational platform for power users to bootstrap their own applications. AVBS abstracts the specifics of runtime interaction with OpenStudio, Modelica, and Spawn of EnergyPlus models behind a unified REST API. Additionally, AVBS provides resources for cloud deployment and scaling to 100s of parallel simulations, a growing library of modular Operational Technology (OT) integrations for emulation of real-world interfaces, and scripts to automate the population of communities of virtual buildings from URBANopt, ResStock and ComStock.

building automation↗

ReEDS Performance Improvement

The Regional Energy Deployment System (ReEDS) is an open-source, spatially explicit, long-term capacity expansion model for the bulk electric power system of the contiguous United States, encompassing multiple scenarios with technological and political assumptions (see https://github.com/NREL/ReEDS-2.0). With the increased needs for capabilities, higher temporal and spatial resolutions to model the evolution of the power system with modern technologies and low-carbon pathways, ReEDS' model solution times have increased significantly from 4-6 hours in 2018 to 18-48+ hours in 2023 . Also, the model size for commonly-run ReEDS scenarios reached 22 and 28 million equations and variables, respectively. These runtimes can be especially challenging under certain scenario settings (e.g., very high temporal or spatial resolution) or with limited computational power. In this presentation, we will discuss several methods we used to improve model runtime, including data preparation, model modification, and solver tuning. The implementation of these methods shrank the model size to 7.2 and 7.3 million equations and variables, respectively. Furthermore, this led to a 77% reduction in the model's run time for commonly-run ReEDS scenarios. We will discuss the process of identifying areas for solve time improvements and how the specific enhancements for the ReEDS model might be applied to other similar large-scale models.

ENERGY PLANNING, POLICY, AND ECONOMY,MATHEMATICS A↗

Fundamental microscopic properties as predictors of large-scale quantities of interest: Validation through grain boundary energy trends

Correlations between fundamental microscopic properties computable from first principles, which we term canonical properties, and complex large-scale quantities of interest (QoIs) provide an avenue to predictive materials discovery. Here, we propose that such correlations can be efficiently discovered through simulations utilizing approximate interatomic potentials (IPs), which serve as an ensemble of “synthetic materials”. As a proof of principle we build a regression model relating canonical properties to the symmetric tilt grain boundary (GB) energy curves in face-centered cubic crystals, characterized by the scaling factor in the universal lattice matching model of Runnels et al. (2016), which we take to be our QoI. Our analysis recovers known correlations of GB energy to other properties and discovers new ones. We also demonstrate, using available density functional theory (DFT) GB energy data, that the regression model constructed from IP data is consistent with DFT results, confirming the assumption that the IPs and DFT belong to same statistical pool and thereby validating the approach. Regression models constructed in this fashion can be used to predict large-scale QoIs based on first-principles data and provide a general method for training IPs for QoIs beyond the scope of first-principles calculations.

36 MATERIALS SCIENCE↗

Approximate Analysis for Interlaminar Stresses in Composite Structures with Thickness Discontinuities

An efficient, approximate analysis for calculating complete three-dimensional stress fields near regions of geometric discontinuities in laminated composite structures is presented. An approximate three-dimensional local analysis is used to determine the detailed local response due to far-field stresses obtained from a global two-dimensional analysis. The stress results from the global analysis are used as traction boundary conditions for the local analysis. A generalized plane deformation assumption is made in the local analysis to reduce the solution domain to two dimensions. This assumption allows out-of-plane deformation to occur. The local analysis is based on the principle of minimum complementary energy and uses statically admissible stress functions that have an assumed through-the-thickness distribution. Examples are presented to illustrate the accuracy and computational efficiency of the local analysis. Comparisons of the results of the present local analysis with the corresponding results obtained from a finite element analysis and from an elasticity solution are presented. These results indicate that the present local analysis predicts the stress field accurately. Computer execution-times are also presented. The demonstrated accuracy and computational efficiency of the analysis make it well suited for parametric and design studies.

Rose, Cheryl A.↗

Operation and Control of Electric Vehicle Charger with Enhanced Dynamic Performance Under Non-Ideal Grid Voltage Condition

This paper presents a three-phase electric vehicle charger connected to the grid, featuring multiple boost converters on the DC side, specifically designed to ensure smooth, oscillation-free power transfer during unsymmetrical voltage sags. Precise control mechanisms are implemented on the boost converter side to regulate both voltage and current on the electric vehicle side, thereby maintaining optimal charging conditions. The control architecture for both the grid-connected and boost converter components is based on the Lyapunov energy function, which is employed to achieve superior dynamic performance and stability. The system's robustness and reliability are demonstrated through its ability to maintain stable operation and efficient power transfer despite fluctuations in grid conditions. Furthermore, the implementation of Lyapunov-based control ensures rapid response and minimal energy loss, enhancing the overall efficiency of the system. To validate the effectiveness of this approach, a comprehensive model of the system was developed and tested using MATLAB/Simulink, with detailed computer simulations conducted across various significant case studies.

DC-DC boost converter↗

Comparison of a Full-Scale and a 1:10 Scale Low-Speed Two-Stroke Marine Engine Using Computational Fluid Dynamics

International marine shipping is a growing component of international trade; a vast majority of all the world’s goods are being transported on large ocean-going vessels. The International Maritime Organization (IMO) introduced the Energy Efficiency Design Index in 2013, a regulatory framework of associated metrics for reducing emissions of CO 2 per tonne-mile from shipping by approximately 10% each decade. Therefore, decarbonizing the maritime sector requires the development of new fuel sources. Because of the extremely large physical size of the internal combustion engines present in shipping vessels, experimental iterative development of the engine and fuel system is cost-prohibitive. Thus, the ability to perform combustion system development in a scaled platform that can be more easily operated and modeled computationally is of interest. To that end, scaling relationships are needed to translate the results from a smaller engine to a larger counterpart. Scaling studies to date have been restricted to low scaling ratios, four-stroke light-duty engines, and under-resolved computational fluid dynamic simulations that likely do not accurately capture the physics of scaling. In this work, computational models of a 1:10 scale and a full-scale two-stroke crosshead low-speed marine engine were created and validated against experiments obtained in a real 1:10 scale engine installed at Oak Ridge National Laboratory. Further, due to the large size of the full-scale engine, the model required large high-performance computing resources to be evaluated. The availability of high-performance computing resources at the Department of Energy’s Leadership Computing Facilities is an enabler of the current work. The results of the small- and large-scale engine simulations were compared to analyze the effectiveness of the appropriate scaling laws under these extreme scaling ratio conditions.

33 ADVANCED PROPULSION SYSTEMS↗

Predicting Accumulations of Ice on Aerodynamic Surfaces

LEWICE is a computer program that predicts the accumulation of ice on two-dimensional aerodynamic surfaces under conditions representative of the flight of an aircraft through an icing cloud. The software first calculates the airflow surrounding the body of interest, then uses the airflow to compute the trajectories of water droplets that impinge on the surface of the body. The droplet trajectories are also used to compute impingement limits and local collection efficiencies, which are used in subsequent ice-growth calculations and are also useful for designing systems to protect against icing. Next, the software predicts the shape of accumulating ice by modeling transfers of mass and energy in small control volumes. The foregoing computations are repeated over several computational time steps until the total icing exposure time is reached. Results of computations by LEWICE have been compared with an extensive database of measured ice shapes obtained from experiments, and have been shown to closely approximate those shapes under most conditions of interest to the aviation community.

Bidwell, Colin↗

Enriching OpenStreetMap network data for transportation applications: Insights into the impact of urban congestion on accessibility

OpenStreetMap (OSM) data is a valuable open-source resource for various transportation, traffic, and planning applications. However, OSM network data lack operating traffic speed information, which is critical for transport planning and operations. Addressing this shortcoming, this study leverages commercial vendor data (to serve as ground truth) with exogenous, open-source variables characterizing local transport infrastructure, land use, and demographic information to predict average congested traffic speeds on OSM networks. Three machine-learning models were tested and estimated for OSM links with and without speed limit information in the Denver metropolitan region. Among these, XGBoost performed best, with mean absolute errors of 3.27 and 3.62 mph for links with and without speed limits, respectively. The developed models accurately predicted traffic speeds for different hours and days of the week compared to ground truth data. Using these predicted speeds, drive accessibility scores were computed for the Denver region for different time periods using the Mobility Energy Productivity (MEP) metric to understand the impact of congestion on energy-efficient accessibility. Results show that congestion-adjusted drive accessibility can be significantly lower compared to accessibility calculated using free flow speeds. Specifically, weekday evening hours saw a 42 % drop in accessibility due to reduced speeds, particularly around downtown Denver. Across the Denver metro region, approximately half as many opportunities and jobs are accessible in under 20 min by car during the evening peak period relative to free flow conditions. These findings underscore the importance of using congestion-adjusted operating speeds rather than speed limits in accessibility calculations, as reliance on speed limits can substantially overestimate energy-efficient drive accessibility in large, car-centric cities susceptible to significant congestion. In conclusion, the methodology presented here could further enrich OSM network data, making them useful for an even broader range of transportation applications.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Unsupervised atomic data mining via multi-kernel graph autoencoders for machine learning force fields

Constructing a chemically diverse dataset while avoiding sampling bias is critical to training efficient and generalizable force fields. However, in computational chemistry and materials science, many common dataset generation techniques are prone to oversampling regions of the potential energy surface. Furthermore, these regions can be difficult to identify and isolate from each other or may not align well with human intuition, making it challenging to systematically remove bias in the dataset. While traditional clustering and pruning (down-sampling) approaches can be useful for this, they can often lead to information loss or a failure to properly identify distinct regions of the potential energy surface due to difficulties associated with the high dimensionality of atomic descriptors. In this work, we introduce the Multi-kernel Edge Attention-based Graph Autoencoder (MEAGraph) model, an unsupervised approach for analyzing atomic datasets. MEAGraph combines multiple linear kernel transformations with attention-based message passing to capture geometric sensitivity and enable effective dataset pruning without relying on labels or extensive training. Demonstrated applications on niobium, tantalum, and iron datasets show that MEAGraph efficiently groups similar atomic environments, allowing for the use of basic pruning techniques for removing sampling bias. This approach provides an effective method for representation learning and clustering that can be used for data analysis, outlier detection, and dataset optimization.

Materials science↗

Scalable training of trustworthy and energy-efficient predictive graph foundation models for atomistic materials modeling: a case study with HydraGNN

We present our work on developing and training scalable, trustworthy, and energy-efficient predictive graph foundation models (GFMs) using HydraGNN, a multi-headed graph convolutional neural network architecture. HydraGNN expands the boundaries of graph neural network (GNN) computations in both training scale and data diversity. It abstracts over message passing algorithms, allowing both reproduction of and comparison across algorithmic innovations that define nearest-neighbor convolution in GNNs. This work discusses a series of optimizations that have allowed scaling up the GFMs training to tens of thousands of GPUs on datasets consisting of hundreds of millions of graphs. Our GFMs use multitask learning (MTL) to simultaneously learn graph-level and node-level properties of atomistic structures, such as energy and atomic forces. Using over 154 million atomistic structures for training, we illustrate the performance of our approach along with the lessons learned on two state-of-the-art US Department of Energy (US-DOE) supercomputers, namely the Perlmutter petascale system at the National Energy Research Scientific Computing Center and the Frontier exascale system at Oak Ridge Leadership Computing Facility. The HydraGNN architecture enables the GFM to achieve near-linear strong scaling performance using more than 2000 GPUs on Perlmutter and 16,000 GPUs on Frontier.

97 MATHEMATICS AND COMPUTING↗

Better, Cheaper, Faster Molecular Dynamics

Recent, revolutionary progress in genomics and structural, molecular and cellular biology has created new opportunities for molecular-level computer simulations of biological systems by providing vast amounts of data that require interpretation. These opportunities are further enhanced by the increasing availability of massively parallel computers. For many problems, the method of choice is classical molecular dynamics (iterative solving of Newton's equations of motion). It focuses on two main objectives. One is to calculate the relative stability of different states of the system. A typical problem that has' such an objective is computer-aided drug design. Another common objective is to describe evolution of the system towards a low energy (possibly the global minimum energy), "native" state. Perhaps the best example of such a problem is protein folding. Both types of problems share the same difficulty. Often, different states of the system are separated by high energy barriers, which implies that transitions between these states are rare events. This, in turn, can greatly impede exploration of phase space. In some instances this can lead to "quasi non-ergodicity", whereby a part of phase space is inaccessible on time scales of the simulation. To overcome this difficulty and to extend molecular dynamics to "biological" time scales (millisecond or longer) new physical formulations and new algorithmic developments are required. To be efficient they should account for natural limitations of multi-processor computer architecture. I will present work along these lines done in my group. In particular, I will focus on a new approach to calculating the free energies (stability) of different states and to overcoming "the curse of rare events". I will also discuss algorithmic improvements to multiple time step methods and to the treatment of slowly decaying, log-ranged, electrostatic effects.

Pohorille, Andrew↗

Optimization of Scrap Melting Using an Electric Arc in Steel Manufacturing

Steel industry is crucial to the national economy and security. Around 67% of crude steel in the U.S is produced in electric arc furnaces (EAF), which is energy intensive. Around 140 EAFs operate in the U.S., consuming about 8.6x10 7 MMBtu/year of electricity. One of major challenges for EAFs includes maximizing the efficiency of the electrical energy provided in the form of electric arcs to melt various scrap mixes. To address this issue, a computational fluid dynamics (CFD) methodology is chosen to analyze scrap melting using the electric arc. Due to complex furnace phenomena and the wide variety of potential scenarios, high performance computing (HPC) is essential to yield comprehensive and detailed CFD analyses and systematic parametric studies for optimized EAF operation. The objectives are to 1) simulate scrap melting using electric arc, 2) evaluate electrode/arc position for optimum scrap melting and 3) establish reduced order model for CFD data-base for fast model calculation.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗