Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “precision”

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 199 records · Page 11

Mixed precision s –step Lanczos and conjugate gradient algorithms

Compared to the classical Lanczos algorithm, the s-step Lanczos variant has the potential to improve performance by asymptotically decreasing the synchronization cost per iteration. However, this comes at a price; despite being mathematically equivalent, the s-step variant may behave quite differently in finite precision, potentially exhibiting greater loss of accuracy and slower convergence relative to the classical algorithm. It has previously been shown that the errors in the s-step version follow the same structure as the errors in the classical algorithm, but are amplified by a factor depending on the square of the condition number of the O(s)-dimensional Krylov bases computed in each outer loop. As the condition number of these s-step bases grows (in some cases very quickly) with s, this limits the s values that can be chosen and thus can limit the attainable performance. In this work, we show that if a select few computations in s-step Lanczos are performed in double the working precision, the error terms then depend only linearly on the conditioning of the s-step bases. This has the potential for drastically improving the numerical behavior of the algorithm with little impact on per-iteration performance. Our numerical experiments demonstrate the improved numerical behavior possible with the mixed precision approach, and also show that this improved behavior extends to mixed precision s-step CG. Here, we present preliminary performance results on NVIDIA V100 GPUs that show that the overhead of extra precision is minimal if one uses precisions implemented in hardware.

97 MATHEMATICS AND COMPUTING↗

Influence of Strain-Gage Balance Characteristics on the Precision Error of the Drag Coefficient

Improvements and results of a new method are presented that computes a pre-test estimate of the precision error of the drag coefficient of a wind tunnel model. The error estimate is defined as the part of the drag coefficient's precision error that is primarily associated with the precision error of the angle of attack measurement and physical characteristics of the chosen strain-gage balance. The method indirectly describes the precision error of the angle of attack measurement by using an assumed balance gage output variation of one microV/V. The physical characteristics of the balance, on the other hand, are described by partial derivatives of the axial and normal forces with respect to the strain-gage outputs. These derivatives can directly be obtained from the data reduction matrix of the balance. The precision error estimate itself is calculated by applying a simple explicit equation that uses the model reference area, the dynamic pressure, the angle of attack, the coefficients of the linear terms of the data reduction matrix, and the electrical output variation of one microvolt per volt as input. Precision errors at constant angle of attack may be visualized as contour plots by plotting them, for example, versus the Mach number and the total pressure. Characteristics of NASA's MC60E balance are used in combination with the reference area of a generic wind tunnel model in order to demonstrate that error estimates are independent of both the balance load format and the units chosen for the description of balance loads, model reference area, and the dynamic pressure. Finally, experimental data from a wind tunnel test of the Ames Check Standard Model in the NASA Ames 11-foot Transonic Wind Tunnel illustrates the application of the method to real-world test data.

precision error↗

Phenomenological study of the interplay between IR-improved DGLAP-CS theory and the precision of an NLO ME matched parton shower MC

We present a phenomenological study of the current status of the application of our approach of exact amplitude-based resummation in quantum field theory to precision QCD calculations, by realistic MC event generator methods, as needed for precision LHC physics. We discuss recent results as they relate to the interplay of the attendant IR-improved DGLAP-CS theory of one of us and the precision of exact NLO matrix-element matched parton shower MC’s in the Herwig6.5 environment as determined by comparison to recent LHC experimental observations on single heavy gauge boson production and decay. The level of agreement between the new theory and the data continues to be a reason for optimism. In the spirit of completeness, we discuss as well other approaches to the same theoretical predictions that we make here from the standpoint of physical precision with an eye toward the (sub-)1% QCD⊗EW total theoretical precision regime for LHC physics. - Highlights: • Using LHC data, we show that IR-improved DGLAP-CS kernels with exact NLO Shower/ME matching improves MC precision. • We discuss other possible approaches in comparison with ours. • We propose experimental tests to discriminate between competing approaches.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

A Novel Phase-Coherent Programmable Clock for High-Precision Arbitrary Waveform Generation Applied to Digital Ion Trap Mass Spectrometry

Digital ion trap (DIT) mass spectrometry requires the ability to precisely and accurately produce waveforms. The quality of the mass spectra produced in terms of resolution and mass accuracy depend on the resolution and precision of the applied waveforms. This publication reveals a novel method for the production of arbitrary waveforms in general and then applies the method to the production of DIT waveforms. Arbitrary waveforms can be created by varying the clock frequency input to a programmable read only memory that is then input to a digital-to-analog converter (DAC). The arbitrary waveform is composed of a defined number of points that are triggered to be written after programmed numbers of clock cycles to define the arbitrary waveform. The novelty introduced here is that the direct digital synthesis (DDS) generated clock frequency can be precisely changed as the arbitrary waveform is written because we have developed a method to rapidly switch the DDS frequency exactly at the end of the output clock cycle allowing exact timing of multiple transitions to produce precise and temporally complex waveforms. Changing the frequency only at the end of the output clock cycle is a phase coherent process that permits precise timing between each point in the arbitrary waveform. The waveform generation technique was demonstrated by creating a prototype that was used to operate a digital ion trap mass spectrometer. The jitter in the phase-coherent DDS TTL output that was used as the frequency variable clock was 20 ps. This jitter represents the realizable limit of precision for waveform generation. The rectangular waveforms used to operate the mass spectrometer were created with counters that increased the jitter to 100 ps. The mass resolution achieved was 5000 at m/z = 414. This resolution corresponds to a jitter of 275 ps assuming DC fluctuations and overshoots in the waveform are insignificant. Resolution should improve with increasing mass because the waveforms have longer periods while the jitter should remain constant. Given the current limit of the variable clock resolution, much better mass resolution should be achievable with future generations of the waveform production system. The agility of the DDS function generators permits the phase-coherent variable clock to be switched at a rate up to 250 MHz. This permits arbitrary waveforms to be produced with much more temporal complexity than previously possible.

47 OTHER INSTRUMENTATION↗

Patterned probes for high precision 4D-STEM bragg measurements

Nanoscale strain mapping by four-dimensional scanning transmission electron microscopy (4D-STEM) relies on determining the precise locations of Bragg-scattered electrons in a sequence of diffraction patterns, a task which is complicated by dynamical scattering, inelastic scattering, and shot noise. These features hinder accurate automated computational detection and position measurement of the diffracted disks, limiting the precision of measurements of local deformation. Here, we investigate the use of patterned probes to improve the precision of strain mapping. We imprint a "bullseye" pattern onto the probe, by using a binary mask in the probe-forming aperture, to improve the robustness of the peak finding algorithm to intensity modulations inside the diffracted disks. We show that this imprinting leads to substantially improved strain-mapping precision at the expense of a slight decrease in spatial resolution. In experiments on an unstrained silicon reference sample, we observe an improvement in strain measurement precision from 2.7% of the reciprocal lattice vectors with standard probes to 0.3% using bullseye probes for a thin sample, and an improvement from 4.7% to 0.8% for a thick sample. Additionally, we use multislice simulations to explore how sample thickness and electron dose limit the attainable accuracy and precision for 4D-STEM strain measurements.

47 OTHER INSTRUMENTATION↗

An Objective and Efficient Method for Assessing the Impact of Reduced-Precision Calculations On Solution Correctness

Recent studies have shown that reducing the precision of floating-point calculations in an atmospheric model can improve the models' computational performance without affecting model fidelity, but code changes are needed to accommodate lower precision, and a small number of calculations might need to retain higher precision to prevent undue round-off error. For complex and computationally expensive systems like the Energy Exascale Earth System Model (E3SM), a method is needed to objectively assess the quality of the lower-precision simulations and to quickly identify problematic code pieces This paper demonstrates that solution correctness can be evaluated through a simple and quantitative error metric based on time step convergence. The proposed test method can unambiguously detect the accumulation of precision error and objectively assess its impact on solution accuracy. Since fast physical processes are known to affect key features of the multi-year mean climate in an atmospheric model, we show that the convergence test applied to short simulations can provide useful information about the impact of reduced precision on a model’s long-term behavior. In contrast, the traditional way of climate model evaluation requires multi-year simulations and involves inspecting many physical quantities for statistical significance in the presence of natural variability. The simplicity, computational efficiency, and objectivity of the proposed method makes it very attractive for high-resolution model development. The method is also expected to be applicable to other models that numerically solve time evolution equations.

97 MATHEMATICS AND COMPUTING↗

Optimizing the Fast Fourier Transform Using Mixed Precision on Tensor Core Hardware

The Fast Fourier Transform is a fundamental tool in scientific and technical computation. The highly parallelizable nature of the algorithm makes it a suitable candidate for GPU acceleration. This paper focuses on exploiting the speedup due to using the half precision multiplication capability of the latest GPUs' tensor core hardware without significantly degrading the precision of the Fourier Transform result. We develop an algorithm that dynamically splits the input single precision dataset into two half precision sets at the lowest level, uses half precision multiplication, and recombines the result at a later step. This work paves the way for using tensor cores for high precision inputs.

Sorna, Anumeena↗

Scaling the memory wall using mixed-precision - HPG-MxP on an exascale-class machine

Mixed-precision algorithms have been proposed as a way for scientific computing to benefit from some of the gains seen for AI on recent high performance computing (HPC) platforms. A few applications dominated by dense matrix operations have seen substantial speedups by utilizing low precision formats such as FP16. However, a majority of scientific simulation applications are memory bandwidth limited. Beyond preliminary studies, the practical gain from using mixed-precision algorithms on a given high-performance computing (HPC) system is largely unclear. The High Performance GMRES Mixed Precision (HPG-MxP) benchmark has been proposed to measure the useful performance of a HPC system on sparse matrix-based mixed-precision applications. In this work, we present an implementation of the HPG-MxP benchmark for an exascale system and describe our algorithm enhancements. We show for the first time a speedup of 1.6x using a combination of double- and single-precision keeping the same residual level on modern GPU-based supercomputers.

Kashi, Aditya [ORNL] (ORCID:0000000325893792)↗

The Precision of Product Consistency Tests with Reference Glasses ARM-1 and SRM 623

An interlaboratory study (ILS) was conducted in 2014 to measure the precision with which the ASTM-International product consistency test method A (PCT-A) can be performed using properly prepared samples of reference glass ARM-1 or SRM 623 and test parameter values near the centers of ranges specified in procedure ASTM C1285. The ILS did not evaluate the full range of test parameter values allowed in the PCT-A method and the results do not represent the precision that should be expected if extreme values of the ranges are used. However, the precision measured in the ILS quantified the precision of tests conducted with mean parameters that were used to estimate the effects of individual parameters. Follow-on tests were conducted to assess the effects of several individual test parameter values on the test response and uncertainties in the derived glass dissolution rates. The effects of particle size distributions in glass samples prepared by wet- and dry-sieving methods, methods to remove fines, and the limiting parameter values for the water-to-glass (w/g) mass ratio and test duration allowed in PCT-A were measured. Test responses for all variations except test duration exceeded the precision measured in the ILS for conducting triplicate PCT-A tests within a single laboratory. The accuracy of glass dissolution rates derived from PCT results benefits from using a tighter control of these test variables than required in ASTM C1285. Comparison with previous analyses of the method shows how use of the relative standard deviation to represent the test precision can be misleading, and that the mean response and standard deviation should both be considered.

36 MATERIALS SCIENCE↗

High-precision measurement of the W boson mass with the CMS experiment at the LHC

In the standard model of particle physics, the masses of the carriers of the weak interaction, the W and Z bosons, are uniquely related. Physics beyond the standard model could change this relationship through the effects of quantum loops of virtual particles, thus making it of great importance to measure these masses with the highest possible precision. Although the mass of the Z boson is known to the remarkable precision of 22 parts per million (2.0 MeV), the W boson mass is known much less precisely, given the difficulty of the measurement. A global fit to electroweak data, used to predict the W boson mass in the standard model, yields an uncertainty of 6 MeV. Reaching a comparable experimental precision would be a sensitive and fundamental test of the standard model. Furthermore, a precision measurement of the W boson mass performed by the CDF Collaboration at the Fermilab Tevatron has challenged the standard model by significantly disagreeing with the prediction of the global electroweak fit and the average of other $m_\mathrm{W}$ measurements. We report the first W boson mass measurement by the CMS Collaboration at the CERN LHC, based on a data sample collected in 2016 at the proton-proton collision energy of 13 TeV. The W boson mass is measured using a large sample of W$\toμν$ events via a highly granular binned maximum likelihood fit to the kinematic properties of the muons produced in the W$^{+}$ and W$^{-}$ boson decays. The significant in situ constraints of theoretical inputs and their corresponding uncertainties, together with an accurate determination of the experimental effects, lead to a precise W boson mass measurement, $m_\mathrm{W} =$ 80$\,$360.2 $\pm$ 9.9 MeV, in agreement with the standard model prediction.

FOS: Physical sciences↗

Annular denuder systems: Precision and applicability to stratospheric sampling

Annular denuder systems (ADS) are designed to sample both gases and particles from ambient air without cross interference. The ADS's were used to detect chemical gradients over rural forest canopies. Accurate and precise methods for the sampling and analysis of low concentrations of gaseous and particulate species over short sampling periods were developed. Typical analytical precision data for nitrate and sulfate are shown. Both sampling and analytical precision data for a 2 hour test using four collocated ADS's are presented. On this particular day, the overall precision for HNO3 was unusually poor. The overall precision for the major constituents of interest is generally about 3 to 5 percent for typical ambient concentrations. The ADS's could potentially be used to determine the concentrations of HNO3 and fine-particle nitrate in the stratosphere, and to simultaneously characterize the chemical composition of stratospheric aerosols. Assuming 50 mb pressure, a 2 hour sampling time, a 50 Lpm sampling rate, and a low blank value, an overal precision of about 15 percent should be possible for 1 ppbv of HNO3 or nitrate, which would include nitric-acid trihydrate (NAT); and pump-driven systems could accommodate more complex systems involving denuders and filter packs in sequence. Alternate coatings could possibly be developed to selectively sample species such as Cl2 and ClNO3. The ADS's might also be useful in laboratory studies of heterogeneous chemical reactions involving NAT aerosols.

Randtke, Stephen J.↗

Towards the GEOSAT Follow-On Precise Orbit Determination Goals of High Accuracy and Near-Real-Time Processing

The US Navy's GEOSAT Follow-On spacecraft (GFO) primary mission objective is to map the oceans using a radar altimeter. Satellite laser ranging data, especially in combination with altimeter crossover data, offer the only means of determining high-quality precise orbits. Two tuned gravity models, PGS7727 and PGS7777b, were created at NASA GSFC for GFO that reduce the predicted radial orbit through degree 70 to 13.7 and 10.0 mm. A macromodel was developed to model the nonconservative forces and the SLR spacecraft measurement offset was adjusted to remove a mean bias. Using these improved models, satellite-ranging data, altimeter crossover data, and Doppler data are used to compute both daily medium precision orbits with a latency of less than 24 hours. Final precise orbits are also computed using these tracking data and exported with a latency of three to four weeks to NOAA for use on the GFO Geophysical Data Records (GDR s). The estimated orbit precision of the daily orbits is between 10 and 20 cm, whereas the precise orbits have a precision of 5 cm.

Lemoine, Frank G.↗

Platform Precision Autopilot Overview and Flight Test Results

The Platform Precision Autopilot is an instrument landing system interfaced autopilot system, developed to enable an aircraft to repeatedly fly nearly the same trajectory hours, days, or weeks later. The Platform Precision Autopilot uses a novel design to interface with a NASA Gulfstream III jet by imitating the output of an instrument landing system approach. This technique minimizes, as much as possible, modifications to the baseline Gulfstream III jet and retains the safety features of the aircraft autopilot. The Platform Precision Autopilot requirement is to fly within a 5-m (16.4-ft) radius tube for distances to 200 km (108 nmi) in the presence of light turbulence for at least 90 percent of the time. This capability allows precise repeat-pass interferometry for the Uninhabited Aerial Vehicle Synthetic Aperture Radar program, whose primary objective is to develop a miniaturized, polarimetric, L-band synthetic aperture radar. Precise navigation is achieved using an accurate differential global positioning system developed by the Jet Propulsion Laboratory. Flight-testing has demonstrated the ability of the Platform Precision Autopilot to control the aircraft within the specified tolerance greater than 90 percent of the time in the presence of aircraft system noise and nonlinearities, constant pilot throttle adjustments, and light turbulence.

Lin, V.↗

Platform Precision Autopilot Overview and Mission Performance

The Platform Precision Autopilot is an instrument landing system-interfaced autopilot system, developed to enable an aircraft to repeatedly fly nearly the same trajectory hours, days, or weeks later. The Platform Precision Autopilot uses a novel design to interface with a NASA Gulfstream III jet by imitating the output of an instrument landing system approach. This technique minimizes, as much as possible, modifications to the baseline Gulfstream III jet and retains the safety features of the aircraft autopilot. The Platform Precision Autopilot requirement is to fly within a 5-m (16.4-ft) radius tube for distances to 200 km (108 nmi) in the presence of light turbulence for at least 90 percent of the time. This capability allows precise repeat-pass interferometry for the Unmanned Aerial Vehicle Synthetic Aperture Radar program, whose primary objective is to develop a miniaturized, polarimetric, L-band synthetic aperture radar. Precise navigation is achieved using an accurate differential global positioning system developed by the Jet Propulsion Laboratory. Flight-testing has demonstrated the ability of the Platform Precision Autopilot to control the aircraft within the specified tolerance greater than 90 percent of the time in the presence of aircraft system noise and nonlinearities, constant pilot throttle adjustments, and light turbulence.

Strovers, Brian K.↗

Apparatus for Precise Indium-Bump Bonding of Microchips

An improved apparatus has been designed and built for use in precise positioning and pressing of a microchip onto a substrate (which could, optionally, be another microchip) for the purpose of indium-bump bonding. The apparatus (see figure) includes the following: A stereomicroscope, A stage for precise positioning of the microchip in rotation angle (theta) about the nominally vertical pressing axis and in translation along two nominally horizontal coordinate axes (x and y), and An actuator system that causes a bonding tip to press the microchip against the substrate with a precisely controlled force. In operation, the microscope and the stage are used to position the microchip under the bonding tip and to align the indium bumps on the chip and the substrate, then the actuator system is used to apply a prescribed bonding force for a prescribed time. The improved apparatus supplants a partly similar prior apparatus that operated with less precision and repeatability, producing inconsistent and unreliable bonds. Results of the use of the prior apparatus included broken microchips, uneven bonds, and bonds characterized, variously, by overcompression or undercompression. In that apparatus, the bonding force was generated and controlled by use of a micrometer head positioned over the center of a spring-loaded scale, and the force was applied to the microchip via the scale, which was equipped for digital readout of the force. The inconsistency of results was attributed to the following causes: It was not possible to control the bonding force with sufficient precision or repeatability. Particularly troublesome was the inability to control the force at levels less than the weight of 150 g. Excessive compliance in the spring-loaded scale, combined with deviations from parallelarity of the substrate and bonding-tip surfaces, gave rise to nonuniformity in the pressure applied to the microchip, thereby generating excessive stresses and deformations in the microchip. In the improved apparatus, the bonding tip and the components that hold the substrate and the microchip are more rigid and precise than in the prior apparatus, so as to ensure less deviation from parallelarity of the bonding-tip and substrate surfaces, thereby ensuring more nearly uniform distribution of bonding force over the area of the microchip. The bonding force is now applied through, and measured by, a load cell that makes it possible to exert finer control over the force. The force can be set at any value between 0 and the weight of 800 g in increments of 0.2 g.

Wild, Larry↗

Precision in Assembled Discrete Lattice Space Structures for Next-Generation ISAM Applications

Robust autonomous robotic assembly of large-scale space structures has been a long-term and challenging goal to enable higher quality in-space communication and science instrumentation. For optical observatory support structures and antenna structures, the challenge is the strict dimensional precision requirements (generally RMS surface error) driven by the operational radiation wavelength. Early theoretical work in the area linked RMS surface error of a reflector support truss plate to the natural passive dynamic mode frequencies of the structure, as well as to the error distribution in dimensions of constituent structural elements. Prior robotic truss assembly demonstrations focused on designing ultra-high precision structural elements, joints, and robotic actuators. Recently, an alternative space structure assembly strategy based on a programmable matter approach (NASA ARMADAS) was demonstrated. This approach uses lattice building blocks (voxels) that are reversibly, mechanically joined into a bulk lattice structure by robots that locomote in and on the structure itself. Such a system can achieve high-level autonomy with low computation, robustly assemble utilizing inexpensive and imprecise robots, and efficiently build structures several orders of magnitude larger than the assembly robots. However, for instrumentation support structure applications, the resulting precision of these building block-based lattice structures is not well studied. Since they are demonstrated with many more assembly units than prior art trusses, it is unclear whether the same precision design approaches apply. In this work, we study in simulation the effects of voxel geometry, error in distribution, assembly resolution (module size), and assembly geometry on the error of both beam and plate lattice structures. While average RMS error of a plate increases with plate size, increasing plate thickness quickly collapses RMS error towards a limit that is on the order of the error of the constituent parts. We validate our models against previously published precision measurements of built systems. Results from this study will guide manufacturing precision requirements, as well as designs, for future robotically assembled structural applications and establish feasibility for different applications.

Christine E Gregg↗

A GPU accelerated mixed-precision Smoothed Particle Hydrodynamics framework with cell-based relative coordinates

Smoothed Particle Hydrodynamics (SPH) is essential for modeling complex large-deformation problems across various applications, requiring significant computational power. A major portion of SPH computation time is dedicated to the Nearest Neighboring Particle Search (NNPS) process. While advanced NNPS algorithms have been developed to enhance SPH efficiency, the potential efficiency gains from modern computation hardware remain underexplored. Here, this study investigates the impact of GPU parallel architecture, low-precision computing on GPUs, and GPU memory management on NNPS efficiency. Our approach employs a GPU-accelerated mixed-precision SPH framework, utilizing low precision float-point 16 (FP16) for NNPS while maintaining high precision for other components. To ensure FP16 accuracy in NNPS, we introduce a Relative Coordinated-based Link List (RCLL) algorithm, storing FP16 relative coordinates of particles within background cells. Our testing results show three significant speedup rounds for CPU-based NNPS algorithms. The first comes from parallel GPU computations, with up to a 1000x efficiency gain. The second is achieved through low-precision GPU computing, where the proposed FP16-based RCLL algorithm offers a 1.5x efficiency improvement over the FP64-based approach on GPUs. By optimizing GPU memory bandwidth utilization, the efficiency of the FP16 RCLL algorithm can be further boosted by 2.7x, as demonstrated in an example with 1 million particles. Our code is released at https://github.com/pnnl/lpNNPS4SPH.

97 MATHEMATICS AND COMPUTING↗

Mixed-precision iterative refinement using tensor cores on GPUs to accelerate solution of linear systems

Double-precision floating-point arithmetic (FP64) has been the de facto standard for engineering and scientific simulations for several decades. Problem complexity and the sheer volume of data coming from various instruments and sensors motivate researchers to mix and match various approaches to optimize compute resources, including different levels of floating-point precision. In recent years, machine learning has motivated hardware support for half-precision floating-point arithmetic. A primary challenge in high-performance computing is to leverage reduced-precision and mixed-precision hardware. We show how the FP16/FP32 Tensor Cores on NVIDIA GPUs can be exploited to accelerate the solution of linear systems of equations Ax = b without sacrificing numerical stability. The techniques we employ include multiprecision LU factorization, the preconditioned generalized minimal residual algorithm (GMRES), and scaling and auto-adaptive rounding to avoid overflow. We also show how to efficiently handle systems with multiple right-hand sides. On the NVIDIA Quadro GV100 (Volta) GPU, we achieve a 4×-5× performance increase and 5× better energy efficiency versus the standard FP64 implementation while maintaining an FP64 level of numerical stability.

GMRES↗