Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Performance 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 1,477 records · Page 82
All sourcesNASA NTRSDOE OSTI
DOE OSTI2015

SOLAR FLARE PREDICTION USING SDO/HMI VECTOR MAGNETIC FIELD DATA WITH A MACHINE-LEARNING ALGORITHM

We attempt to forecast M- and X-class solar flares using a machine-learning algorithm, called support vector machine (SVM), and four years of data from the Solar Dynamics Observatory's Helioseismic and Magnetic Imager, the first instrument to continuously map the full-disk photospheric vector magnetic field from space. Most flare forecasting efforts described in the literature use either line-of-sight magnetograms or a relatively small number of ground-based vector magnetograms. This is the first time a large data set of vector magnetograms has been used to forecast solar flares. We build a catalog of flaring and non-flaring active regions sampled from a database of 2071 active regions, comprised of 1.5 million active region patches of vector magnetic field data, and characterize each active region by 25 parameters. We then train and test the machine-learning algorithm and we estimate its performances using forecast verification metrics with an emphasis on the true skill statistic (TSS). We obtain relatively high TSS scores and overall predictive abilities. We surmise that this is partly due to fine-tuning the SVM for this purpose and also to an advantageous set of features that can only be calculated from vector magnetic field data. We also apply a feature selection algorithm to determine which of our 25 features are useful for discriminating between flaring and non-flaring active regions and conclude that only a handful are needed for good predictive abilities.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2026

Neural Scaling Laws for Jet Generation

Recently observed empirical scaling laws describe the performance of foundation-type models as three independent key quantities -- dataset size, compute, and model parameters -- are modified. Extracting these scaling laws informs the training of large complex models for which the tuning of hyperparameters in traditional ways is not feasible. This work for the first time explores if scaling laws can also be observed for the task of particle jet generation -- both relevant as a pre-training objective for foundation models and as in-situ simulation by itself. We indeed replicate the key logarithmic scaling law behavior for model-size scaling. Beyond studying the next token prediction validation loss of the generative model, we also study the sliced Wasserstein distance of five physical quantities that are not immediately available to the model during training. Our study shows that this quantity is monotonically related to the next token prediction validation loss, meaning that this loss is indeed a good proxy for the physics performance. For the scaling with dataset size and compute, we observe substantially weaker scaling behavior of both the loss and the sliced Wasserstein distance. We analyze this behavior by introducing the concept of a learnable window, and argue that autoregressive next token prediction on jet constituents exhibits comparatively rapid saturation relative to language-model studies. We discuss possible origins of this behavior, including the stochastic nature of QCD radiation and differences between generative and supervised learning tasks in collider physics.

Amram, Oz [Fermilab]↗
DOE OSTI2022

Language models for the prediction of SARS-CoV-2 inhibitors

The COVID-19 pandemic highlights the need for computational tools to automate and accelerate drug design for novel protein targets. We leverage deep learning language models to generate and score drug candidates based on predicted protein binding affinity. We pre-trained a deep learning language model (BERT) on ∼9.6 billion molecules and achieved peak performance of 603 petaflops in mixed precision. Our work reduces pre-training time from days to hours, compared to previous efforts with this architecture, while also increasing the dataset size by nearly an order of magnitude. For scoring, we fine-tuned the language model using an assembled set of thousands of protein targets with binding affinity data and searched for inhibitors of specific protein targets, SARS-CoV-2 Mpro and PLpro. We utilized a genetic algorithm approach for finding optimal candidates using the generation and scoring capabilities of the language model. Our generalizable models accelerate the identification of inhibitors for emerging therapeutic targets.

Blanchard, Andrew E.↗
DOE OSTI2018

Quantifying the inter- and intra-granular stresses and their effects on deformation behaviors for the polycrystalline hexagonal beryllium by neutron diffraction

Highlights: • The inter- and intra-granular stresses are quantified for the polycrystalline hexagonal metal beryllium. • Composite behavior of the internal stresses is revealed in the polycrystalline beryllium. • Those internal stresses show the obvious grain-orientation dependence. • Those stresses can be significantly tuned by the temperature and strain rate effects. • The dislocation slip and twinning are quantified together with the inter- and intra-granular stored energies. - Abstract: The internal stresses developing at different length scales play the important role in materials performance. This importance has been well recognized for the cubic metals, however, not for the hexagonal close-packed (hcp) ones. The present work investigates the intergranular and intragranular stresses in the polycrystalline hexagonal beryllium by neutron diffraction. The correlation between those internal stresses and the deformation mechanism is revealed, and the strain rate and temperature effects on those stresses are presented. A composite behavior of intergranular and intragranular stresses is observed, which can be significantly tuned by the temperature and strain rate effects. The intergranular stress and stored energy are large for the soft-oriented grains (SOGs), while the intragranular stress and stored energy are large for the hard-oriented ones (HOGs). Both of those stresses are significantly high for the dynamic strain rate while they are relatively low for the elevated temperature. The large internal stresses in the dynamically compressed specimen should be attributed to the large number of twins and high density of dislocations. Different HOGs-SOGs configurations are proposed on the light of those observations, which might facilitate the properties controlling and extend the application scope for the hcp metal beryllium.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Highly efficient non-fullerene organic solar cells enabled by a delayed processing method using a non-halogenated solvent

The use of non-halogenated solvents is one of the important requirements for large-scale industry production of polymer solar cells (PSCs). However, the current high-performance PSCs are mostly obtained using halogenated solvents to ensure good solubility and morphology. In this work, we report high-performance PSCs processed using a halogen-free solvent (toluene), reaching an efficiency of 17.33%, which, to the best of our knowledge, is the highest value reported for PSCs processed from non-halogenated solvents. Importantly, we show a novel aggregation or “aging” process of the non-fullerene acceptor in the toluene solvent, which can be used to fine-tune the crystal size of the non-fullerene acceptor. As a result, we achieve an optimized nano-fibril like morphology of PM6:BTP-BO-4Cl by delaying the processing time. Furthermore ,this is conducive to charge dissociation and transport, and thus greatly improved device performance. This method is expected to be investigated further and applied to other PSC systems toward precise morphology optimization and performance improvements in the future.

36 MATERIALS SCIENCE↗
DOE OSTI2019

Enhancing the Performance of a Fused-Ring Electron Acceptor by Unidirectional Extension

The unidirectional extension of a smaller fused-ring system into a larger one in a single direction will increase the conjugation length, allowing a fine-tuning of electronic properties. Here, we designed and synthesized a unidirectionally extended fused-8-ring-based nonfullerene acceptor, AOIC, and a bidirectionally extended fused-11-ring electron acceptor, IUIC2, and compared these with the parent fused-5-ring electron acceptor, F5IC. They share the same electron-accepting groups and alkylphenyl side chains but have different fused-ring electron-donating units. Core extension from 5 to 11 rings up-shifts the energy levels, red shifts the absorption spectra, and reduces bandgaps. The unidirectionally extended AOIC has the highest mobility (2.1 × 10$^{-3}$ cm$^2$ V$^{-1}$ s$^{-1}$) relative to the parent F5IC (1.0 × 10-3 cm2 V-1 s$^{-1}$) and the bidirectionally extended IUIC2 (4.7 × 10$^{-4}$ cm$^2$ V-1 s$^{-1}$). Upon blending with the donor PTB7-Th, AOIC-based organic photovoltaic cells show an efficiency of 13.7%, much better than that of F5IC-based cells (5.61%) and IUIC2-based cells (4.48%).

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2014

Boron-tuning transition temperature of vanadium dioxide from rutile to monoclinic phase

The effect of the doped boron on the phase transition temperature between the monoclinic phase and the rutile phase of VO{sub 2} has been studied by performing first-principles calculations. It is found that the phase transition temperature decreases linearly with increasing the doping level of B in each system, no matter where the B atom is in the crystal. More importantly, the descent of the transition temperature is predicted to be as large as 83 K/at. % B, indicating that the boron concentration of only 0.5% can cause the phase transition at room temperature. These findings provide a new routine of modulating the phase transition of VO{sub 2} and pave a way for the practicality of VO{sub 2} as an energy-efficient green material.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2017

Progress on the Magnetic Performance of Planar Superconducting Undulators

One of the primary goals of the superconducting undulator (SCU) program at the Advanced Photon Source (APS) is to achieve a high quality undulator magnetic field without the need for magnetic shimming to tune the device. Over the course of two years, two SCUs were designed, manufactured, assembled, and tested at the APS. Both SCUs were one meter in length with a period of 1.8 cm. After magnetic measurements of the first undulator were completed, several design changes were made in order to improve the quality of the undulator magnetic field. The design modifications were implemented during construction and assembly of the second SCU. The details of the design modifications along with a comparison of the magnetic measurement results will be described.

Kasa, M.↗
DOE OSTI2026

Demonstration of gold nanorod systems for enhanced total efficiency: Experimental and numerical analysis

This study investigates the photothermal performance of gold nanorods engineered to exhibit longitudinal plasmon resonances at 695 nm, 780 nm, and 970 nm. The work combines synthesis, structural characterization, extinction measurements, numerical modeling, and controlled temperature experiments to quantify how nanorod geometry, resonance tuning, concentration, and chamber shape jointly influence heat generation. Transmission electron microscopy confirms that increasing nanorod aspect ratio systematically shifts the longitudinal plasmon peak toward the near-infrared region. Extinction measurements show strong agreement with theoretical predictions, validating the numerical model across two independent datasets. Three chamber geometries were tested under laser excitation at 640 nm, 808 nm, and 980 nm: an ascending stepped base, a flat base, and a descending stepped base. Without nanorods, the ascending geometry produced the highest efficiency due to enhanced natural convection. After introducing gold nanorods, all geometries exhibited substantial thermal enhancement, with total efficiencies exceeding 20%. The strongest improvement was obtained for nanorods resonant at 780 nm with a mass concentration of 4.6 mg/mL implemented on the descending stepped-base geometry. This performance resulted from the combined effect of spectral overlapping with the 808 nm laser, the highest nanorod concentration, and localized heat accumulation that intensified buoyancy-driven flow. The findings demonstrate that total efficiency is governed by a synergistic interplay between optical resonance, nanoparticle concentration, and macroscopic chamber design, revealing the system-level coupling between nanoscale plasmonic absorption and macroscale heat-transfer phenomena. The results provide a validated framework for tuning nanoscale plasmonic absorbers and optimizing thermal systems for applications requiring efficient light-to-heat conversion.

77 NANOSCIENCE AND NANOTECHNOLOGY↗
DOE OSTI2018

Tuning Electronic Structure of Single Layer MoS 2 through Defect and Interface Engineering

Transition-metal dichalcogenides (TMDs) have emerged in recent years as a special group of two-dimensional materials and have attracted tremendous attention. Among these TMD materials, molybdenum disulfide (MoS 2 ) has shown promising applications in electronics, photonics, energy, and electrochemistry. In particular, the defects in MoS 2 play an essential role in altering the electronic, magnetic, optical, and catalytic properties of MoS 2 , presenting a useful way to engineer the performance of MoS 2 . The mechanisms by which lattice defects affect the MoS 2 properties are unsettled. In this work, we reveal systematically how lattice defects and substrate interface affect MoS 2 electronic structure. We fabricated single-layer MoS 2 by chemical vapor deposition and then transferred onto Au, single-layer graphene, hexagonal boron nitride, and CeO 2 as substrates and created defects in MoS 2 by ion irradiation. Here, we assessed how these defects and substrates affect the electronic structure of MoS 2 by performing X-ray photoelectron spectroscopy, Raman and photoluminescence spectroscopies, and scanning tunneling microscopy/spectroscopy measurements. Molecular dynamics and first-principles based simulations allowed us to conclude the predominant lattice defects upon ion irradiation and associate those with the experimentally obtained electronic structure. We found that the substrates can tune the electronic energy levels in MoS 2 due to charge transfer at the interface. Furthermore, the reduction state of CeO 2 as an oxide substrate affects the interface charge transfer with MoS 2 . The irradiated MoS 2 had a faster hydrogen evolution kinetics compared to the as-prepared MoS 2 , demonstrating the concept of defect controlled reactivity in this phase. Our findings provide effective probes for energy band and defects in MoS 2 and show the importance of defect engineering in tuning the functionalities of MoS 2 and other TMDs in electronics, optoelectronics, and electrochemistry.

42 ENGINEERING↗
DOE OSTI2020

Tailoring a Three-Phase Microenvironment for High-Performance Oxygen Reduction Reaction in Proton Exchange Membrane Fuel Cells

Despite tremendous progress in catalyst development for rate-limiting cathodic oxygen reduction reaction (ORR), reducing Pt usage while meeting performance requirements in practical proton exchange membrane fuel cells (PEMFCs) remains a challenge. The ORR in PEMFCs occurs at a catalyst–electrolyte–gas three-phase interface. A desirable interface should exhibit highly active and available catalytic sites, as well as allow efficient oxygen and proton feeding to the catalytic sites and timely removal of water to avoid interface flooding. In this work, we report the design of a three-phase microenvironment in PEFMCs, showing that carbon surface chemistry can be tuned to modulate its interaction with the ionomers and create favorable transport paths for rapid delivery of both reactants and products. With such an elaborate interfacial design, for the first time we have demonstrated PEMFCs with all key ORR catalyst performance metrics, including mass activity, rated power, and durability, surpassing the US Department of Energy targets.

25 ENERGY STORAGE↗
DOE OSTI2019

Numerical study and test of the APS linac transverse deflecting cavity

The transverse deflecting cavity can be used to transform particle distributions in the 6D phase space, which makes it a promising component in phase space beam diagnostics and beam manipulations. In the Advanced Photon Source (APS) Linac, a LOLA-type traveling wave deflecting cavity was installed for the diagnostics of beam characteristics, such as the bunch length, bunch temporal profile, time-dependent energy spread and slice (time-correlated) transverse emittance. In this paper, this deflecting cavity is modeled and analyzed with the numerical method. The effects of the center cell and coupler cell dimensions on the performance of the whole structure are studied, which shows the coupler cell radius has a dominant effect over the coupler cell length and slot width on the global reflection coefficient and field flatness. Important RF parameters, such as S-11, field flatness and phase advance etc., of the LOLA-II cavity in tuning are calculated and discussed. After optimization, the field flatness of the cavity is 2.1%, the phase advance is 119.66 degrees degrees with a standard deviation smaller than 0.5 degrees and the bandwidth is 15.738 MHz when VSWR < 1.1 (i.e. the S-11, < -26 dB). Bead-pull measurements and RF conditioning have been also performed and discussed. Finally, the initial cavity commissioning with electron beam of 375 MeV is reported, and compared with simulation.

coupler↗
DOE OSTI2018

BaCe 0.25 Mn 0.75 O 3–δ —a promising perovskite-type oxide for solar thermochemical hydrogen production

Solar-thermal based hydrogen production technologies employing two-step metal oxide water-splitting cycles are emerging as a viable approach to renewable and sustainable solar fuels. However, materials innovations that overcome thermodynamic constraints native to the current class of solar-thermal water splitting oxides are required to increase solar utilization and process efficiency. Lowering oxide thermal reduction temperature while maintaining high water-splitting favorability are important ways to enhance such performance metrics. Recent attention to perovskite-type oxides as an alternative to ceria, which is widely viewed as the state-of-the art redox material, is driven by demonstrated thermodynamic and structural tuning derived through engineered composition. Here we discuss the unique properties of BaCe 0.25 Mn 0.75 O 3 (BCM) within the context of thermochemical water splitting materials. Firstly, BCM is a novel example of a line compound with B-site substitution of Mn by Ce. It also exhibits a polymorph phase transition during thermal reduction and yields nearly 3× more H 2 than ceria when reduced at lower temperature (1350 °C). More importantly, BCM exhibits faster oxidation kinetics and higher water-splitting favorability than Sr x La 1–x Mn y Al 1–y O 3 (x, y = 0.4, 0.6), which is a well-studied and popular Mn-based perovskite formulation. Furthermore, the unique properties manifested by BCM through engineered composition offer new pathways towards unlocking higher performing materials for solar thermochemical water splitting.

14 SOLAR ENERGY↗
DOE OSTI2016

Regulation of flow through a T-Shaped open cavity by temperature dependent P, PI, and PID controllers

P (proportional), PI (proportional-integral), and PID (proportional-integral-derivative) controllers are popular means of controlling industrial processes. Due to superior response, accuracy, and stable performance, PID controllers are mostly used in control systems. This paper presents a mathematical model and subsequent response analysis regarding regulation of flow in mixed convection through a T-shaped open cavity by temperature dependent controllers. The T-shaped cavity has cold top and hot bottom walls, while air is flowing through the inlet at surrounding temperature. The inflow is regulated by a controlled gate which operates according to the signal received from the controller. Values of proportional gain (k{sub p}), integral gain (k{sub i}), and derivative gain (k{sub d}) are varied to obtain the desired system response and to ensure a stable system with fastest response. At first, only P controller is used and eventually PI and finally PID control scheme is applied for controller tuning. Tuning of different controllers (P, PI, and PID) are carried out systematically based on the reference temperature which is continuously monitored at a certain location inside the cavity. It is found that PID controller performs better than P or PI controller.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2019

Strain-Induced Structural Deformation Study of 2D Mo x W (1- x ) S 2

The potential ability of tuning properties and its possible applications in the fields of optoelectronics and/or flexible electronics, has increased the demand for 2D alloys in recent times. Understanding the mechanical performance of 2D materials under extreme conditions, such as strain, stress, and fracture is essential for the reliable electronic devices based on these structures. Here, combined molecular dynamics (MD) simulations and in situ Raman spectroscopic techniques are used to study the mechanical performance of a 2D alloy system, MoxW (1- x ) S 2 . It is determined that W substitution in MoS 2 causes solid-solution strengthening and increase in the Young's modulus values. Higher W content decreases failure strain for MoS 2 . Due to spatially resolved Raman spectroscopy and MD simulations results, a detailed model to explain failure mechanisms in MoxW (1- x ) S 2 alloys is proposed.

36 MATERIALS SCIENCE↗
DOE OSTI2019

Impact of Surface Composition of SrTiO 3 Catalysts for Oxidative Coupling of Methane

Abstract Oxidative coupling of methane (OCM) to C 2 hydrocarbons (C 2 H 6 and C 2 H 4 ) have regained much attention due to the shale gas revolution. Perovskite catalysts have shown promising activity and selectivity to C 2 hydrocarbons. Here, we investigate the effect of surface reconstruction (leading to different surface compositions) of perovskites on the OCM by using SrTiO 3 (STO) as a model catalyst. Different surface densities of Sr (25–96 %) were attained via various treatments of STO. Low energy ion scattering (LEIS) and UV‐Raman results are in good agreement on the surface and subsurface composition of the reconstructed STO. From H 2 ‐TPR, the same H 2 consumption of STO samples allows relating their catalytic performances with surface acid‐base properties (quantified by NH 3 ‐/CO 2 ‐TPD). At 600–800 °C, the surface Sr enrichment was found to enhance CH 4 conversion, C 2 selectivity and the ratio C 2 H 4 /C 2 H 6 up to Sr/(Sr+Ti) of 0.66 and then levels off. Furthermore, the relative concentration of basic sites, base/(base+acid), is found as a better descriptor for STO catalytic performances. This work shows the clear correlation between surface reconstruction, relative basicity/acidity and OCM catalytic performance over perovskite catalysts. The trends here are similar to those for CH 4 combustion over the reconstructed STO in our recent work. Overall, we suggest that tuning surface reconstruction/composition of perovskites can be an effective approach to control CH 4 activation and conversions.

Bai, Lei↗
DOE OSTI2024

A machine learning approach for particle accelerator errant beam prediction using spatial phase deviation

Particle accelerators are extremely complex systems that are expected to operate on high availability. Predicting impending failures only by utilizing data collected from diagnostic equipment already on board can help operators to avoid installing expensive sensors, unscheduled downtime and associated costs. For this purpose we explore the predictive power of Machine Learning algorithms to detect faulty beams prior to the failure. In this study, we propose a Machine Learning approach to model mapping from a pair of sensors located across the accelerator. While the model is trained to represent normal operation, we evaluate the predictive performance on known faulty beam pulses. We also investigate the model performance on unseen data through k-fold cross-validation. Then we recap the analysis with a neural architecture search and hyperparameter optimization study to fine tune our initial model. In conclusion, this paper will also introduce a sustainable framework that can standardize Machine Learning workflow applied to particle accelerators.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗