Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Tuning”

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
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

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↗
DOE OSTI2024

High Capacity Step-Shaped Hydrogen Adsorption in Robust, Pore-Gating Zeolitic Imidazolate Frameworks (Final Technical Report)

The use of porous adsorbents for the densification of H 2 under relatively mild pressures and temperatures has been long pursued through various avenues of synthetic organic frameworks, namely metal–organic frameworks. However, the overwhelming majority of these materials are macroscopically rigid, exhibiting adsorption–desorption profiles that require appreciable energetic input for their full payload capacity to be stored and subsequently delivered. This reduced “deliverable” capacity is a core inefficiency in the use of porous adsorbents for H 2 densification. Our core mission is to determine a material-based strategy that fundamentally alters the adsorption¬–desorption profile in a manner that obviates these deleterious energetic requirements. Towards this, we have focused on metal–organic frameworks that exhibit “cooperative flexibility”, wherein the material undergoes reversible macroscopic phase changes in response to changes to external temperature and/or adsorbate pressure (i.e., concentration). Where these reversible switches occur between states of disparate enough accessible porosity, cooperatively flexible frameworks can exhibit “step-shaped” adsorption–desorption profiles that enable use of the delivery gaseous payload without the aforementioned additional energy requirements. However, before our work, such the observation of this phenomenon with very weakly adsorbing H 2 had not been observed with the necessary profile and P/T conditions desired for H 2 storage and delivery. During this project we have leveraged a deep understanding of the phase change in family of cooperatively flexible metal–organic frameworks to systematically alter their adsorption–desorption profiles to achieve step-shaped adsorption and desorption of H 2 in a precise P/T regime that enables a high deliverable capacity. Specifically, the thermodynamics of the adsorption/desorption induced phase changes in the baseline framework CdIF-13 (sod–Cd(benzimidazolate) 2 ) were tuned using the mixed-linker, or multivariate, approach to framework derivatization. Follow this, we have pursued a similar approach to tuning the adsorption–desorption profile of a second framework family, MIL-53(Al), which contains only Al(III) and theoretically exhibits a much higher deliverable capacity. In total, our work demonstrates that by first understanding the structural consequences of phase changes in these materials, their behavior can be systematically altered through ligand substitutions, tuning the pressure and temperature conditions at which H 2 is adsorbed and desorbed. Thus, enabling high deliverable capacities of H 2 with minimized energetic requirements.

08 HYDROGEN↗
DOE OSTI2025

Study of Quadrupole Fringe Fields in the Interaction Region of the Hadron Storage Ring of the Electron Ion Collider

Fringe fields in quadrupole magnets are usually neglected in studies of beam dynamics at accelerators. However, the extreme optical parameters present in the final focus of a collider such as the Electron–Ion Collider (EIC) may give rise to effects that should not be overlooked. The calculation of quadrupole fringe fields presented in this study follows the procedure outlined in Ref. [1], specialized to the case of a straight reference orbit (i.e., with no dipole field component). A right-handed Cartesian coordinate system is employed, with the $z$-axis aligned with the quadrupole axis and $x$ and $y$ denoting the horizontal and vertical transverse coordinates, respectively. The magnetic quadrupole field gradient, $G = \partial B_y / \partial x$, transitions from its peak value inside the quadrupole—where it is nearly independent of the longitudinal coordinate $z$—to zero at some distance beyond the magnet edge. Consequently, $G$ is treated as a function of $z$. The region over which this variation occurs is defined as the quadrupole fringe field region. The study begins with the development of a description of the magnetic field in the fringe region using a power-series expansion in the transverse coordinates $x$and $y$, consistent with the longitudinally varying gradient. A model for the $z$-dependence of the gradient is then proposed and adjusted to reproduce magnetic field data obtained from three-dimensional field calculations. To evaluate the impact of the fringe fields on beam dynamics, the corresponding vector potential is derived and incorporated into a Hamiltonian formulation of particle motion. The significance of the fringe fields is quantified by calculating the amplitude-dependent tune shift from the Hamiltonian. Using linear beam optics parameters of the Hadron Storage Ring (HSR) of the EIC, the tune shift due to the fringe fields of all quadrupole magnets in the IR-6 interaction region is evaluated. Finally, the resulting tune shifts are compared with those arising from other nonlinear field components present in the HSR.

43 PARTICLE ACCELERATORS↗
DOE OSTI2026

Hofmann Stability Charts Revisited for PIP-II: From Classical Theory to Assumption-Free and ML-Driven Maps

The Hofmann stability chart remains a standard for visualizing parametric resonances in space-charge–dominated linacs, but its use typically relies on non-oscillatory Vlasov dispersion relations with simplifying assumptions (continuous focusing, KV phase space, linear optics, limited transverse–longitudinal coupling). We revisit the chart for the PIP-II linac along three tracks. (1) We reproduce the conventional maps in the (νz/νx, νx/ν0x) plane for relevant εz/εx, providing a validated reference. (2) We remove key assumptions by deriving stability surfaces directly from multi-particle tracking with realistic lattice discreteness, RF defocusing, solenoid/quad optics, and bunched-beam dynamics; local tunes and early-time growth rates are estimated from envelope oscillations and projected to the same coordinates. These assumption-reduced maps recover the canonical stopbands while revealing shifts and broadenings driven by tune modulation, non-KV distributions, and transverse–longitudinal coupling at PIP-II intensities. (3) We train a compact machine-learning surrogate that emulates the growth surface from zero-current optics, tune depression, emittance ratio, bunching factor, and selected lattice descriptors, enabling rapid scans and online working-point selection. We compare the three representations on representative PIP-II sections and discuss implications for commissioning guard bands, resonance avoidance, and routine operations.

Pathak, Abhishek [Fermilab] (ORCID:000000021704208↗
DOE OSTI2024

Optimizing Distributed Training on Frontier for Large Language Models

Large language models (LLMs) have demonstrated remarkable success as foundational models, benefiting various downstream applications through fine-tuning. Loss scaling studies have demonstrated the superior performance of larger LLMs compared to their smaller counterparts. Nevertheless, training LLMs with billions of parameters poses significant challenges and requires considerable computational resources. For example, training a one trillion parameter GPT-style model on 20 trillion tokens requires a staggering 120 million exaflops. This research explores efficient distributed training strategies to extract this computation from Frontier, the world's first exascale supercomputer. We enable and investigate various model and data parallel training techniques, such as tensor parallelism, pipeline parallelism, and sharded data parallelism, to facilitate training a trillion-parameter model on Frontier. We empirically assess these techniques and their associated parameters to determine their impact on memory footprint, communication latency, and GPU's computational efficiency. We analyze the complex interplay among these techniques and find a strategy to combine them to achieve high throughput through hyperparameter tuning. We have identified efficient strategies for training large LLMs of varying sizes through empirical analysis and hyperparameter tuning. For 22 Billion, 175 Billion, and 1 Trillion parameters, we achieved GPU throughputs of 38.38%, 36.14%, and 31.96%, respectively. For the training of the 175 Billion parameter model and the 1 Trillion parameter model, we achieved 100% weak scaling efficiency on 1024 and 3072 Mi250X GPUs, respectively. We also achieved strong scaling efficiencies of 89% and 87% for these two models. We trained these models only tens of iterations instead of training till completion.

Yin, Junqi↗
DOE OSTI2025

End-to-end deep learning pipeline for real-time Bragg peak segmentation: from training to large-scale deployment

X-ray crystallography reconstruction, which transforms discrete X-ray diffraction patterns into three-dimensional molecular structures, relies critically on accurate Bragg peak finding for structure determination. As X-ray free electron laser (XFEL) facilities advance toward MHz data rates (1 million images per second), traditional peak finding algorithms that require manual parameter tuning or exhaustive grid searches across multiple experiments become increasingly impractical. While deep learning approaches offer promising solutions, their deployment in high-throughput environments presents significant challenges in automated dataset labeling, model scalability, edge deployment efficiency, and distributed inference capabilities. We present an end-to-end deep learning pipeline with three key components: (1) a data engine that combines traditional algorithms with our peak matching algorithm to generate high-quality training data at scale, (2) a modular architecture that scales from a few million to hundreds of million parameters, enabling us to train large expert-level models offline while deploying smaller, distilled models at the edge, and (3) a decoupled producer-consumer architecture that separates specialized data source layer from model inference, enabling flexible deployment across diverse computing environments. Using this integrated approach, our pipeline achieves accuracy comparable to traditional methods tuned by human experts while eliminating the need for experiment-specific parameter tuning. Although current throughput requires optimization for MHz facilities, our system's scalable architecture and demonstrated model compression capabilities provide a foundation for future high-throughput XFEL deployments.

Wang, Cong↗
DOE OSTI2024

ZnO-Au Hybrid Metamaterial Thin Films with Tunable Optical Properties

ZnO-Au nanocomposite thin films have been previously reported as hybrid metamaterials with unique optical properties such as plasmonic resonance properties and hyperbolic behaviors. In this study, Au composition in the ZnO-Au nanocomposites has been effectively tuned by target composition variation and thus resulted in microstructure and optical property tuning. Specifically, all the ZnO-Au nanocomposite thin films grown through the pulsed laser deposition (PLD) method show obvious vertically aligned nanocomposite (VAN) structure with the Au nanopillars embedded in the ZnO matrix. Moreover, the average diameter of Au nanopillars increases as Au concentration increases, which also leads to the redshifts in the surface plasmon resonance (SPR) wavelength and changes in the hyperbolic behaviors of the films. As a whole, this work discusses how strain-driven tuning of optical properties and microstructure resulted through a novel Au concentration variation approach which has not been previously attempted in the ZnO-Au thin film system. These highly ordered films present great promise in the areas of sensing, waveguides, and nanophotonics to name a few.

Bhatt, Nirali A.↗
DOE OSTI2024

Impact of the Exciter and Governor Parameters on Forced Oscillations

In recent years, the frequency of forced oscillation events due to control system malfunctions or improper parameter settings has increased. Tuning the parameters of exciters and governor models is crucial for maintaining power system stability. Traditional simulation studies typically involve small transient disturbances or step changes to find optimal parameter sets, but existing optimization algorithms often fall short in fine-tuning for forced oscillations. Identifying the sensitive parameters within these control models is essential for ensuring stability during large, sustained disturbances. This study focuses on identifying these critical exciter and governor model parameters by analyzing their influence on sustained forced oscillations. Using Kundur’s two-area system, we analyze common exciter models such as SCRX, ESST1A, and AC7B, along with governor models like GAST, HYGOV, and GGOV1, utilizing PSS®E software version 34. Sustained forced oscillations are injected at generator-1 of area-1, with individual parameter changes dynamically simulated. By considering a local oscillation frequency of 1.4 Hz and an inter-area oscillation mode of 0.25 Hz, we analyze the impact of each parameter change on the magnitude and frequency of forced oscillations as well as on active and reactive power outputs. This novel approach highlights the most influential parameters of each tested model—such as exciter, governor, and turbine gains, as well as time constant parameters—on the impact of forced oscillations. Based on our findings, the sensitive parameters of each tested model are ranked. These would provide valuable insights for industry operators to fine-tune control settings during oscillation events, ultimately enhancing system stability.

42 ENGINEERING↗
DOE OSTI2021

Biodegradable surfactants and related compositions, methods and systems

Biodegradable surfactants are described, in which an amphiphilic heteroatom containing hydrocarbon optionally comprising at least one counterion (Z), and related compositions, methods and systems. Biodegradable surfactant described herein has an aHLB value in accordance with equation (1): aHLB=20*Gh/(Gh−Gt) (1) wherein Gh is the Group Number of a hydrophilic head portion of the biodegradable surfactant optionally comprising the at least one counterion (Z), and Gt is the Group Number of a hydrophobic tail portion of the biodegradable surfactant. A biodegradable surfactant in the sense of the disclosure can be tuned to a set hydrophilic-lipophilic balance (aHLB) by selectively modifying at least one tuning moiety of the biodegradable surfactants to provide tuned biodegradable surfactants having an increase or decrease in their adjusted hydrophilic-lipophilic balance (aHLB).

Lyman, Mathew Gerald↗
DOE OSTI2024

Axion Dark Matter EXperiment 2A Cavity Characterization

We cooled the ADMX 2A cavities down to 5K and found the Q-factors for various tunings, which were about a factor of five larger than the Q-factors from the previous cooldown. We also found that the tuning mechanism with the piezo motor and gearbox is not currently equipped to handle cryogenic temperatures. We created an experimental warm bead pull setup using piezo motors, an alumina bead, a pulley system, and a distance calibrating script. We found that the bead pull method can successfully show us the reflection-induced electric field mode structure. This technique allowed us to differentiate between different TE and TM modes near mode crossings. However, some bead pull sweeps produced noisy or asymmetrical results, possibly due to asymmetries in the tuning rod orientation or antenna. Future iterations of the bead pull method can help us estimate the cavity form factor and maximize our scan range near mode crossings.

Maglione, D.↗
DOE OSTI2024

SCITUNE: Aligning Large Language Models with Human-Curated Scientific Multimodal Instructions

Instruction finetuning is a popular paradigm to align large language models (LLM) with human intent. Despite its popularity, this idea is less explored in improving the LLMs to align existing foundation models with scientific disciplines, concepts and goals. In this work, we present SciTune as a tuning framework to improve the ability of LLMs to follow scientific multimodal instructions. To test our methodology, we use a human-generated scientific instruction tuning dataset and train a large multimodal model LLaMA-SciTune that connects a vision encoder and LLM for science-focused visual and language understanding. LLaMA-SciTune significantly outperforms the state-of-the-art models in the generated figure types and captions in multiple scientific multimodal benchmarks. In comparison to the models that are fine-tuned with machine generated data only, LLaMA-SciTune surpasses human performance on average and in many sub-categories on the ScienceQA benchmark.

• Artificial intelligence (AI) / machine learning ↗
DOE OSTI2024

A tunable dielectric resonator for axion searches at 11 GHz

In the context of axion search with haloscopes, tunable cavity resonators with high quality factor and high effective volume at frequencies above about 8 GHz are central for probing the axion-photon coupling with the required sensitivity to reach the QCD axion models. Higher order modes in dielectrically-loaded cavities allow for higher effective volumes and larger quality factors compared to basic cylindrical cavities, but a proper cavity frequency tuning mechanism to probe broad axion mass ranges is yet not available. In this work, we report about the design and construction of a tunable prototype of a single-shell dielectric resonator with its axion-sensitive pseudo-TM$_{030}$ high-order mode at about 11 GHz frequency. A clamshell tuning method previously tested for empty cylindrical resonators has been perfected for this geometry through simulations and in situ tests conducted at cryogenic temperature. Tuning is accomplished in a range of about 2 $\%$ the central frequency, without significantly impacting the quality factor of about 175000. The experimental results presented in this work demonstrate the tunability of this type of resonator, definitely confirming its applicability to high frequency axion searches.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2025

Mechanical Actuation Mechanism Design for Cryogenic Microwave Cavities in Axion Dark Matter Searches

Resonant Microwave Cavities are placed in a strong magnetic field at cryogenic temperatures (~.1K) in order to try to convert axions into detectable photons. A major challenge is to tune the cavity to allow no interference with the desired environment used for the experiment, which enables the system to operate near absolute zero and maintain superconductivity, thereby reducing noise as much as possible. My project focuses on developing and testing a low-thermal-conductivity gear reduction system with a 1000:1 ratio for cavity tuning. This gear system could significantly improve the thermal and electromagnetic interference of the tuning mechanism, helping ADMX scan for axions more accurately.

Ortega, Giancarlo [Marquette U.; Fermilab]↗
DOE OSTI2026

Learning to Trigger: Reinforcement Learning at the Large Hadron Collider

High-throughput scientific facilities such as the Large Hadron Collider depend on real-time event filtering (\textit{triggering}) under tight constraints on bandwidth, latency, and storage. In practice, trigger menus are largely static and hand-tuned and can become suboptimal as detector conditions, pileup, and background composition drift over time. We cast online threshold tuning as a sequential decision-making problem: a reinforcement learning agent ingests streaming summaries of recent rates and signal-sensitive features and updates trigger thresholds to maximize signal efficiency while tracking a target background rate within a tolerance band. We adapt Group-Filtered Policy Optimization (GFPO) to streaming control and introduce two variants (GFPO-F, GFPO-FR) that enforce background rate feasibility during training. On a benchmark that emulates realistic collider operation, we study two representative triggers: a total transverse energy ($H_{T}$) trigger sensitive to pileup variation, and an anomaly-detection (AD) trigger based on reconstruction loss for rare or non-standard signatures. On Monte Carlo streams, our agent increases the fraction of in-tolerance time intervals by 48% ($H_T$) and 28% (AD), with a cumulative gain of up to 2% in signal efficiency on those in-tolerance intervals. Transferring from simulation to \emph{real} collision data (CMS Run 283408), the same agent, without fine-tuning, achieves a 56% ($H_T$) and 28% (AD) in-tolerance improvement over baselines, with further signal-efficiency gain on both triggers. To our knowledge, this is the \emph{first} demonstration of RL-based trigger control on real Large Hadron Collider collision data. Code is available at https://github.com/Zixind/GFPO_LHC (see repo for details).

Ding, Zixin [Chicago U.]↗
DOE OSTI2023

Experimental investigation of advanced turbine control strategies and load-mitigation measures with a model-scale floating offshore wind turbine system

To advance the control co-design of offshore wind energy systems, the authors perform basin-scale experiments with a fully instrumented and actuated floating offshore wind turbine model. The model consists of a 1:70 scale performance-matched model of the International Energy Agency Wind Technology Collaboration Programme 15-MW reference turbine atop the VolturnUS-S semisubmersible platform. The Reference OpenSource Controller provides real-time blade pitch and generator torque control. Here we aim to develop an open data set for the validation of numerical models in predicting the influence of turbine control and load-mitigation measures. For this purpose, we measure the effects of advanced turbine control features, including peak shaving and floating feedback as well as hull-based control using tuned mass dampers on the system. Overall, the results demonstrate measurable and consistent influences from the control and load-mitigation measures, thus confirming the usefulness as a validation data set. Peak shaving attenuates the response to wind turbulence at near-rated wind speed. Floating feedback reduces the load and platform pitch motion associated with the negative damping induced by blade pitch control. The tuned mass dampers also attenuate the system response near the targeted frequencies under suitable conditions. We also identify detrimental side effects of each load-mitigation measure.

17 WIND ENERGY↗
DOE OSTI2025

AI-assisted transport of radioactive ion beams

Beams of radioactive heavy ions allow researchers to study rare and unstable atomic nuclei, shedding light into the internal structure of exotic nuclei and on how chemical elements are formed in stars. However, the extraction and transport of radioactive beams rely on time-consuming expert-driven tuning methods, where hundreds of parameters are manually optimized. Here, in this study, we introduce a system that employs Artificial Intelligence (AI), specifically utilizing Bayesian Optimization, to assist in the transport process of radioactive beams. We apply our methodology to real-life scenarios showing advantages when compared with standard tuning methods. This AI-assisted approach can be extended to other radioactive beam facilities around the world to improve operational efficiency and enhance scientific output.

43 PARTICLE ACCELERATORS↗
DOE OSTI2025

Controlling beam trajectory and transport in a tapered helical undulator

In this paper we present a detailed discussion of the helical undulator system developed for high extraction efficiency experiments in the tapered-enhanced stimulated superradiant amplification regime. The design is based on permanent magnet technology and comprises two Halbach arrays orthogonally oriented and shifted by 90° with respect to each other. When used in low energy beamlines for THz generation, the electron beam trajectory and transport are particularly sensitive to the undulator off-axis fields so that it becomes important to complement on-axis field measurements with analysis and tuning of the higher-order field components. Here we describe how the pulsed wire measurement technique can be effectively used to retrieve on- and off-axis magnetic field characteristics of the undulator. A simple two-dipole model is developed to guide the final adjustments to the permanent magnet positions along the array to tune the quadrupole and sextupole components of the field.

free-electron lasers↗
DOE OSTI2022

Improving OFDR Distributed Fiber Sensing by Fibers With Enhanced Rayleigh Backscattering and Image Processing

This article investigates using optical fibers with enhanced backscattering profiles to improve distributed fiber sensor performance and reduce instrumentation costs. Using a femtosecond (fs) laser direct writing technique, the Rayleigh backscattering profile of a standard telecom fiber was enhanced by more than 40 dB to improve the signal-to-noise ratio (SNR) for optical frequency domain reflectometry (OFDR). The enhanced backscattering signals enable effectively distributed strain measurements using a low-cost tunable laser (TL). Median filtering is applied to denoise cross correlation results to further improve measurement outcomes. Results presented in this article show that a TL with a 1-nm tuning range, which is far less than the tuning range used in commercial OFDR interrogators, can perform effectively distributed strain measurements using sensing fibers with enhanced backscattering profiles. The sensing fiber with over 40-dB backscattering enhancement achieved 4.8-cm spatial resolution in strain measurements with a root mean square accuracy of less than 2.70 με when 10–50 με were exerted to the sensing fiber. In conclusion, results presented in this article reveal both the potential and limitations of sensing fibers with enhanced backscattering for OFDR-based distributed fiber sensors.

42 ENGINEERING↗