Search NASA⌕ Search

SEARCH · Search NASA

Results for “parameter selection”

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 1,225 records · Page 68

NEUTRAL HYDROGEN OPTICAL DEPTH NEAR STAR-FORMING GALAXIES AT z Almost-Equal-To 2.4 IN THE KECK BARYONIC STRUCTURE SURVEY

We study the interface between galaxies and the intergalactic medium by measuring the absorption by neutral hydrogen in the vicinity of star-forming galaxies at z Almost-Equal-To 2.4. Our sample consists of 679 rest-frame UV-selected galaxies with spectroscopic redshifts that have impact parameters <2 (proper) Mpc to the line of sight of one of the 15 bright, background QSOs and that fall within the redshift range of its Ly{alpha} forest. We present the first two-dimensional maps of the absorption around galaxies, plotting the median Ly{alpha} pixel optical depth as a function of transverse and line-of-sight separation from galaxies. The Ly{alpha} optical depths are measured using an automatic algorithm that takes advantage of all available Lyman series lines. The median optical depth, and hence the median density of atomic hydrogen, drops by more than an order of magnitude around 100 kpc, which is similar to the virial radius of the halos thought to host the galaxies. The median remains enhanced, at the >3{sigma} level, out to at least 2.8 Mpc (i.e., >9 comoving Mpc), but the scatter at a given distance is large compared with the median excess optical depth, suggesting that the gas is clumpy. Within 100 (200) kpc, and over {+-}165 km s{sup -1}, the covering fraction of gas with Ly{alpha} optical depth greater than unity is 100{sup +0}{sub -32}% (66% {+-} 16%). Absorbers with {tau}{sub Ly{alpha}} > 0.1 are typically closer to galaxies than random. The mean galaxy overdensity around absorbers increases with the optical depth and also as the length scale over which the galaxy overdensity is evaluated is decreased. Absorbers with {tau}{sub Ly{alpha}} {approx} 1 reside in regions where the galaxy number density is close to the cosmic mean on scales {>=}0.25 Mpc. We clearly detect two types of redshift space anisotropies. On scales <200 km s{sup -1}, or <1 Mpc, the absorption is stronger along the line of sight than in the transverse direction. This 'finger of God' effect may be due to redshift errors, but is probably dominated by gas motions within or very close to the halos. On the other hand, on scales of 1.4-2.0 Mpc the absorption is compressed along the line of sight (with >3{sigma} significance), an effect that we attribute to large-scale infall (i.e., the Kaiser effect).

79 ASTRONOMY AND ASTROPHYSICS↗

RW Aur A FROM THE X-WIND POINT OF VIEW: GENERAL FEATURES

In this paper, the RW Aur A microjet is studied from the point of view of X-wind models. The archived Hubble Space Telescope/STIS spectra of the optical forbidden lines [O I], [S II], and [N II] from RW Aur A, taken in Cycle 8 with seven parallel slits along the jet axis, spaced at 0.''07 apart, were analyzed. Images, position-velocity diagrams, and line ratios among the species were constructed, and compared with synthetic observations generated by selected solutions of the X-wind. Prominent features arising in a steady-state X-wind could be identified within the convolved images and position-velocity diagrams, including FWHM and high-velocity peaks on both of the redshifted and blueshifted jets. The well-known asymmetric velocity profiles of the opposite jets were built into the selected models. We discuss model selections within the existing uncertainties of the stellar parameters and inclination angle of the system. In this framework, the mass-loss rates that were inferred to be decreasing along the jet axis in the literature are the results of slowly decreasing excitation conditions and electron density profiles. Despite the apparent asymmetry in the terminal velocities, line intensities and mass-loss rates, the average linear momenta from the opposite sides of the jet are actually balanced. These previously hard-to-explain features of the asymmetric RW Aur A jet system can now be interpreted in a different but self-consistent manner within the X-wind framework.

79 ASTRONOMY AND ASTROPHYSICS↗

ULTRAVIOLET COLOR-COLOR RELATION OF EARLY-TYPE GALAXIES AT 0.05 < z < 0.12

We present the ultraviolet (UV) color-color relation of early-type galaxies (ETGs) in the nearby universe (0.05 < z < 0.12) to investigate the properties of hot stellar populations responsible for the UV excess (UVX). The initial sample of ETGs is selected by the spectroscopic redshift and the morphology parameter from the Sloan Digital Sky Survey (SDSS) Data Release 7, and then cross-matched with the Galaxy Evolution Explorer (GALEX) far-UV (FUV) and near-UV (NUV) GR6 data. The cross-matched ETG sample is further classified by their emission line characteristics in the optical spectra into quiescent, star-forming, and active galactic nucleus categories. Contaminations from early-type spiral galaxies, mergers, and morphologically disturbed galaxies are removed by visual inspection. By drawing the FUV-NUV (as a measure of UV spectral shape) versus FUV - r (as a measure of UVX strength) diagram for the final sample of {approx}3700 quiescent ETGs, we find that the 'old and dead' ETGs consist of a well-defined sequence in UV colors, the 'UV red sequence', so that the stronger UVX galaxies should have a harder UV spectral shape systematically. However, the observed UV spectral slope is too steep to be reproduced by the canonical stellar population models in which the UV flux is mainly controlled by age or metallicity parameters. Moreover, 2 mag of color spreads both in FUV-NUV and FUV - r appear to be ubiquitous among any subsets in distance or luminosity. This implies that the UVX in ETGs could be driven by yet another parameter which might be even more influential than age or metallicity.

79 ASTRONOMY AND ASTROPHYSICS↗

The role of radio loudness in Eigenvector 1 and the Baldwin Effect of [O iii ] λ5007

ABSTRACT The radio emission is an important observable of quasars, but its relationship to Eigenvector 1 and the [O iii] λ5007 Baldwin Effect is not fully understood. We explore these issues based on a sample of 1800 quasars taken from the Sloan Digital Sky Survey Data Release 7. We employ a new approach of selecting subsamples in the plane of fundamental physical parameters of the black hole mass and Eddington ratio, so as to reduce variables and complexity in analyses. Based on these subsamples, we investigate the relationship between radio loudness R and Eigenvector 1 and find that radio loudness is correlated with [O iii] λ5007 emission, but has no clear relationship with optical Fe ii emission, which indicates that the radio power is probably not a driver of Eigenvector 1 but merely a secondary process. In addition, we also investigate the impact of radio loudness on the Baldwin Effect of [O iii] λ5007. We find that when the radio loudness is not strong (log R < 2), the Baldwin Effect of [O iii] λ5007 is clear, while in the samples of extreme radio loudness (log R ≥ 2), the Baldwin Effect of [O iii] λ5007 becomes weaker or even disappears. We suggest that both radio loudness and luminosity influence the relative strength of [O iii].

79 ASTRONOMY AND ASTROPHYSICS↗

Anchor-based optimization of energy density functionals

A new anchor-based optimization method of defining energy density functionals (EDFs) is proposed. In this approach, the optimization of the parameters of EDFs is carried out for a selected set of spherical anchor nuclei, the physical observables of which are modified by the correction function, which takes into account the global performance of EDFs. It is shown that the use of this approach leads to a substantial improvement in the global description of binding energies for several classes of covariant EDFs. Here, the computational cost of defining a new functional within this approach is drastically lower as compared with the one for the optimization which includes the global experimental data on spherical, transitional, and deformed nuclei in the fitting protocol.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Approximating lattice similarity

A method is proposed for choosing unit cells for a group of crystals so that they all appear as nearly similar as possible to a selected cell. Related unit cells with varying cell parameters or indexed with different lattice centering can be accommodated.

36 MATERIALS SCIENCE↗

Effect of process route on powder three-dimensional-printing of metal powders

The purpose of this study is to investigate the effect of two unique processing routes (solvent jetting (SJ) and binder jetting (BJ)), on the green density of printed stainless steel 316L (SS316L) and Nickel (Ni) powders. In the SJ processing route, a solvent is jetted unto the powder/binder mixture to selectively activate the binder, layer by layer. In the BJ processing route, a solution of the binder mixture is jetted onto the powder bed to selectively bind powder particles. The effects of printing parameters such as layer height, roller speed, shaker speed and nozzle temperature on the green density of printed components are investigated and compared for both processing routes. Results show that layer height and nozzle temperature affect the relative density of the printed compact for both processing routes. Slightly higher relative densities were achieved via the SJ route, with the overall highest relative density being 42.7% at 100 µ m layer height and 70% nozzle temperature for the SS316L components and 43.7% at 150 µ m layer height and 90% nozzle temperature for the Ni components, respectively. Results also show an increase in the final sintered relative density with an increase in green (printed) relative density of the solvent jetted SS316L components, with the highest relative density being 87.2%. The paper studies the influence of printing parameters on the green density of printed SS316L and Ni samples in an unprecedented effort to provide a comparative understanding of the process-property relationships in BJ and SJ of SS316L and Ni components to the additive manufacturing research community.

Engineering↗

A Visual Designer of Layer‐wise Relevance Propagation Models

Abstract Layer‐wise Relevance Propagation (LRP) is an emerging and widely‐used method for interpreting the prediction results of convolutional neural networks (CNN). LRP developers often select and employ different relevance backpropagation rules and parameters, to compute relevance scores on input images. However, there exists no obvious solution to define a “best” LRP model. A satisfied model is highly reliant on pertinent images and designers' goals. We develop a visual model designer, named as VisLRPDesigner, to overcome the challenges in the design and use of LRP models. Various LRP rules are unified into an integrated framework with an intuitive workflow of parameter setup. VisLRPDesigner thus allows users to interactively configure and compare LRP models. It also facilitates relevance‐based visual analysis with two important functions: relevance‐based pixel flipping and neuron ablation. Several use cases illustrate the benefits of VisLRPDesigner. The usability and limitation of the visual designer is evaluated by LRP users.

Huang, Xinyi↗

SU-D-17A-03: 5D Respiratory Motion Model Based Iterative Reconstruction Method for 4D Cone-Beam CT

Purpose: The purpose of this work is to develop a new iterative reconstruction method for 4D cone-beam CT (CBCT) based on a published time-independent 5D respiratory motion model. The proposed method will offer a single high-resolution image at a user-selected breathing phase and the 5D motion model parameters, which could be used to generate the breathing pattern during the CT acquisition. Methods: 5D respiratory motion model was proposed for accurately modeling the motion of lung and lung tumor tissues. 4D images are then parameterized by a reference image, measured breathing amplitude, breathing rate, two time-independent vector fields that describe the 5D model parameters, and a scalar field that describes the change in HU as a function of breathing amplitude. In contrast with the traditional method of reconstructing multiple temporal image phases to reduce respiratory artifact, 5D model based method simplify the problem into the reconstruction of a single reference image and the 5D motion model parameters. The reconstruction formulation of the reference image and scalar and vector fields is a nonlinear least-square optimization problem that consists of solving the reference image and fields alternately, in which the reference image is regularized with the total variation sparsity transform and the vector fields are solved through linearizations regularized by the H1 norm. 2D lung simulations were performed in this proof-of-concept study. Results: The breathing amplitude, its rate, and the corresponding scalar and vector fields were generated from a patient case. Compared with filtered backprojection method and sparsity regularized iterative method for the phase-by-phase reconstruction, the proposed 5D motion model based method yielded improved image quality. Conclusion: Based on 5D respiratory motion model, we have developed a new iterative reconstruction method for 4D CBCT that has the potential for improving image quality while providing needed on-table motion information.

60 APPLIED LIFE SCIENCES↗

Analysis of Advanced Fuel Assemblies and Core Designs for the Current and Next Generations of LWRs

The objective of the project is to design and analyze advanced fuel assemblies for use in current and future light water reactors and to assess their ability to reduce the inventory of transuranic elements, while preserving operational safety. The reprocessing of spent nuclear fuel can delay or avoid the need for a second geological repository in the US. Current light water reactor fuel assembly designs under investigation could reduce the plutonium inventory of reprocessed fuel. Nevertheless, these designs are not effective in stabilizing or reducing the inventory of minor actinides. In the course of this project, we developed and analyzed advanced fuel assembly designs with improved thermal transmutation capability regarding transuranic elements and especially minor actinides. These designs will be intended for use in thermal spectrum (e.g., current and future fleet of light water reactors in the US). We investigated various fuel types, namely high burn-up advanced mixed oxides and inert matrix fuels, in various geometrical designs that are compliant with the core internals of current and future light water reactors. Neutronic/thermal hydraulic effects were included. Transmutation efficiency and safety parameters were used to rank and down-select the various designs.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

An Integrated Supercritical System for Efficient Produced Water Treatment and Power Generation

The main goal of this project was to evaluate the feasibility of an integrated supercritical (SC) water desalination and power generation system for treatment of produced waters from CO 2 sequestration, oil production, and coal-bed methane recovery. This system may offer a transformative approach to generating power from coal or natural gas (NG) and purifying saline water or produced water with high total dissolved solids (TDS) in a zero-liquid discharge plant. The closed-loop steam cycle of conventional power plants was replaced with an open-loop steam cycle that heated the pretreated produced water above the SC point of water (374 °C and 221 bar) to precipitate and separate the salts dissolved in the water. After separation of the precipitated salt, the SC steam required further treatment to reduce the salt content from ~100 mg/L (ppm) to <1 ppm to make it suitable for use in turbines for power generation. A SC membrane distillation unit was used to polish the SC steam, which could be sent through a series of turbines for power generation, and the steam exiting the low-pressure turbine could be condensed by cooling water to produce pure water. Project tasks included a process simulation and a techno-economic evaluation of the integrated system; design and assembly of SC salt precipitation and membrane distillation systems; development and characterization of advanced carbon membranes for SC membrane distillation; and pretreatment, desalination, and purification of different produced water samples with salt concentrations of 30,000–200,000 ppm to a high-purity (<1 ppm) level. Process simulations were performed for 550 MW SC coal-fired or NG baseline cases and various scenarios for the proposed integrated SC system using produced water as the water source and coal or NG as the fuel source. Simulation results indicated that the net plant efficiency loss of the proposed SC cogeneration process would be in the range of <1% to ~4%, in which NG and lower TDS produced water cases would have higher efficiencies. The proposed process for coal-fired cases could provide all water needs for the power generation process and generate additional pure water for beneficial use. A techno-economic analysis was performed to estimate the cost of produced water treatment (dollars /kgal of treated water) by conventional treatment processes compared with the proposed cogeneration process. The cost estimation for the proposed cogeneration system included the cost of pretreatment, the cost of power loss to integrate the water desalination process into the power generation system, and the cost of plant modification or retrofitting. The cost of produced water treatment was calculated by assuming three different scenarios in which converting a baseline power plant to the proposed cogeneration plant would increase the cost of the boiler and feed water system by 10%, 50%, and 100%. Results indicated that the cost of produced water treatment, even with a 100% cost increase, was in the range of $8.6–11.6 dollars per kgal, which is about half of the treatment cost by conventional methods. The cost of produced water treatment by the proposed cogeneration system could be further reduced or a considerable income could be generated if the values of saltwater and freshwater products are considered. Overall, the techno-economic analysis results indicated that the proposed integrated SC system could potentially treat produced water at a cost significantly lower than the cost of produced water treatment by conventional technologies. To evaluate the feasibility of the proposed treatment system at a laboratory scale, a SC salt precipitation and membrane distillation system was designed and fabricated to treat a simulated brine or pretreated produced water by precipitating salts and further polishing the SC steam containing ~100 ppm salts to a high-purity steam. Through baseline tests with NaCl solutions, with concentrations ranging from 30,000 to 200,000 ppm, it was demonstrated that concentrated feed solutions could be treated by precipitating salt at SC conditions to a TDS level of 130 ppm. Produced water samples were collected from a potential CO 2 sequestration site, a coal-bed methane site, and an oilfield in Illinois. Representative samples were characterized for various water quality parameters, including pH; turbidity; conductivity; and concentrations of selected anions, cations, alkalinity, dissolved organic carbon, total suspended solids, TDS, and total petroleum hydrocarbons. The produced water samples were first treated by conventional pretreatment methods to remove residual oil, dissolved organic matter, and scale-forming species. Pretreatment methods included coagulation, sedimentation, filtration, activated carbon adsorption, and ion-exchange adsorption. Pretreated produced waters were treated at SC conditions in which salts were precipitated. Analysis of treated samples revealed that the concentration of Na (the main cation) in produced waters was reduced dramatically from 14,000–47,400 ppm in the feed to 21–80 ppm in treated waters. Concentrations of other main cations (Ca, Mg, Sr, K) also decreased significantly from initial values of ~1,000–22,000 ppm to ~1 ppm or less. However, a significant increase in the concentrations of Fe, Ni, Cu, and Co was observed, mainly caused by leaching of these elements from the system (vessels, filters, tubing), which is manufactured from alloys containing Fe, Ni, Cu, and Co. To polish the SC steam containing a salt content of ~100 ppm to a high-purity level of <1 ppm, a SC membrane distillation treatment using carbon membranes was employed. Different types of self-standing or supported carbon membranes were developed from graphite, graphene, carbon nanotubes, carbon nanofibers, and pyrolytic carbon. Screening tests revealed that only a limited number of developed carbon membranes were suitable for the purification of steam at SC conditions. Membranes that were either fragile and lacked mechanical strength or had large pores and showed limited salt rejection (<20%) were found to be unsuitable for the purification of SC steam. Promising carbon membranes that were successfully tested for purifying SC steam included two graphite membranes (i.e., AF-5, a porous graphite material, and FGS, a flexible graphite sheet) and a carbon nanotube membrane grown on a Hastelloy substrate and then further modified using a proprietary method (i.e., membrane C). The AF-5 membrane exhibited a stable water flux of ~50 kg/m 2 ∙h and ~40% salt rejection. The FGS membrane had a high salt rejection of 95%–98% but a lower water flux of ~15 kg/m 2 ∙h. Membrane C showed an extremely high flux of 200-800 kg/m 2 ∙h, which is one order of magnitude higher than the water flux of conventional membranes. Membrane C also exhibited 87%–93% salt rejection. Overall, a two-stage or multistage polishing treatment by these membranes could lower the salt content of the SC steam from initial values of ~100–200 ppm to the steam quality of <1 ppm or to parts per billion levels required for the power cycle.

20 FOSSIL-FUELED POWER PLANTS↗

Intelligent electrical outlet for collective load control

Various technologies described herein pertain to an electrical outlet that autonomously manages loads in a microgrid. The electrical outlet can provide autonomous load control in response to variations in electrical power generation supply in the microgrid. The electrical outlet includes a receptacle, a sensor operably coupled to the receptacle, and an actuator configured to selectively actuate the receptacle. The sensor measures electrical parameters at the receptacle. Further, a processor autonomously controls the actuator based at least in part on the electrical parameters measured at the receptacle, electrical parameters from one or more disparate electrical outlets in the microgrid, and a supply of generated electric power in the microgrid at a given time.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Quant-CT: Segmenting and Quantifying Computed Tomography

Quant-CT is currently a plugin to ImageJ, designed as a Java-class that provides control mechanism for the user to choose volumes of interest within porous material, followed by the selection of image subsamples for automated tuning of parameters for filters and classifiers, and finally measurement of material geometry, porosity, and visualization. Denoising is mandatory before any image interpretation, and we implemented a new 3D java code that performs bilateral filtering of data. Segmentation of the dense material is essential before any quantifications about geological sample structure, and we invented new schemes to deal with over segmentation when using statistical region merging algorithm to pull out grains that compose imaged material. It make uses of ImageJ API and other standard and thirty-party APIs. Quant-CT conception started in 2011 under Scidac-e sponsor, and details of the first prototype were documented in publications below. While it is used right now for microtomography images, it can potentially be used by anybody with 3D image data obtained by experiment or produced by simulation.

Daniela Ushizima, Wes Bethel↗

Plasticity models of material variability based on uncertainty quantification techniques

The advent of fabrication techniques like additive manufacturing has focused attention on the considerable variability of material response due to defects and other micro-structural aspects. This variability motivates the development of an enhanced design methodology that incorporates inherent material variability to provide robust predictions of performance. In this work, we develop plasticity models capable of representing the distribution of mechanical responses observed in experiments using traditional plasticity models of the mean response and recently developed uncertainty quantification (UQ) techniques. Lastly, we demonstrate that the new method provides predictive realizations that are superior to more traditional ones, and how these UQ techniques can be used in model selection and assessing the quality of calibrated physical parameters.

36 MATERIALS SCIENCE↗

Synchrophasor measurement method for power systems

A synchrophasor measurement method for a device configured to take synchronized measurements in a power system is provided. The synchrophasor measurement method includes receiving global positioning system (GPS)-synchronized samples of a signal sensed by the device from the power system; determining a level of distortion of the signal; selecting, based on the level of distortion, a computation method, the computation method being one of an improved zero-crossing (IZC) method and an enhanced phase-lock-loop (EPLL) method; performing the selected computation method to determine at least one parameter of the signal at a reporting frequency, which is at least twice a line frequency of the power system; and outputting, at the reporting frequency, the at least one parameter to an operator of the power system to allow the operator to perform at least one of a monitoring and a controlling of at least one element of the power system.

Yao, Wenxuan↗

A Visual Designer of Layer-wise Relevance Propagation Models

Layer-wise Relevance Propagation (LRP) is an emerging and widely-used method for interpreting the prediction results of convolutional neural networks (CNN). LRP developers often select and employ different relevance backpropagation rules and parameters, to compute relevance scores on input images. However, there exists no obvious solution to define a "best" LRP model. A satisfied model is highly reliant on pertinent images and designers' goals. We develop a visual model designer, named as VisLRPDesigner, to overcome the challenges in the design and use of LRP models. Various LRP rules are unified into an integrated framework with an intuitive workflow of parameter setup. VisLRPDesigner thus allows users to interactively configure and compare LRP models. It also facilitates relevance-based visual analysis with two important functions: relevance-based pixel flipping and neuron ablation. Several use cases illustrate the benefits of VisLRPDesigner. The usability and limitation of the visual designer is evaluated by LRP users.

97 MATHEMATICS AND COMPUTING↗

ML-ACCEPT: Machine-Learning-enhanced Automated Circuit Configuration and Evaluation of Power Converters

This project, ML-ACCEPT, took a significant step toward achieving the first objective of the DIFFERENTIATE program: to help engineers to more rapidly and cost-effectively consider a wider range of more novel concepts before selecting an engineering-optimal architecture for high-fidelity detailed design and evaluation. Compared to existing methods, the ML-ACCEPT project investigated innovative technologies to make the design of power converters more cost-effective and time-efficient by i) integrating recent breakthroughs in ML, power electronics, simulation software, and optimization to research, develop, and developed a suite of ML-enhanced hypothesis generation tools for power converter design; and ii) facilitating the integration of the proposed software tools into existing power-converter design work-flows. Some existing tools for power-converter design have a certain level of intelligence to aid the design process, but to our knowledge none of them has ML-enhanced capability of automating the electrical circuit design process by selecting, configuring, and tailoring the individual components that enable available resources (e.g., high-voltage DC power) to be converted to the desired output (e.g., lower-voltage DC power with a desired voltage ripple), subject to application-specific (e.g., plasma generation and automotive applications) thermal and packaging considerations. The state-of-the-art circuit design of power converters is still heavily reliant on human experts to select the optimal topology and search for design parameters with human’s experience and intuitions, which can be very time-consuming, inefficient, and labor intensive. The team is composed of University of Michigan-Dearborn (UM-Dearborn), Lawrence Livermore National Laboratory (LLNL), Oak Ridge National Laboratory (ORNL), Modelon Inc., and lighthouse customers.

42 ENGINEERING↗

Efficient, direct compilation of SU(N) operations into SNAP & Displacement gates

We present a function which connects the parameter of a previously published short sequence of selective number-dependent arbitrary phase (SNAP) and displacement gates acting on a qudit encoded into the Fock states of a superconducting cavity, $V_k(\alpha)=D(\alpha)R_\pi(k)D(-2\alpha)R_\pi(k)D(\alpha)$ to the angle of the Givens rotation $G(\theta)$ on levels $|k\rangle,|k+1\rangle$ that sequence approximates, namely $\alpha=\Phi(\theta) = \frac{\theta}{4\sqrt{k+1}}$. Previous publications left the determination of an appropriate $\alpha$ to numerical optimization at compile time. The map $\Phi$ gives us the ability to compile directly any $d$-dimensional unitary into a sequence of SNAP and displacement gates in $O(d^3)$ complex floating point operations with low constant prefactor, avoiding the need for numerical optimization. Numerical studies demonstrate that the infidelity of the generated gate sequence $V_k$ per Givens rotation $G$ scales as approximately $O(\theta^6)$. We find numerically that the error on compiled circuits can be made arbitrarily small by breaking each rotation into $m$$\theta/m$ rotations, with the full $d\times d$ unitary infidelity scaling as approximately $O(m^{-4})$. This represents a significant reduction in the computational effort to compile qudit unitaries either to SNAP and displacement gates or to generate them via direct low-level pulse optimization via optimal control.

Job, Joshua↗