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 1,693 records · Page 94
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Ionomer Optimization for Water Uptake and Swelling in Anion Exchange Membrane Electrolyzer: Hydrogen Evolution Electrode

Green hydrogen produced through anion exchange membrane water electrolysis is a promising, low-cost chemical storage solution for intermittent renewable energy sources. Low-temperature electrolysis using anion exchange membranes (AEM) combines the benefits of established water electrolysis technologies based on alkaline electrolysis and proton exchange membrane electrolysis. The anion conductive ionomers (ACI) used in the AEM electrolyzer (AEMEL) electrodes has been investigated. The ACI serves two primary purposes: (i) facilitate hydroxide conduction between the catalyst and bulk electrolyte and (ii) bind the catalyst to the porous transport layer and membrane. High ion exchange capacity (IEC) ACIs are desired, however, high IEC can cause excessive water uptake (WU) and detrimental ACI swelling. Proper water management is a key factor in obtaining maximum performance in AEM-based devices. In this study, a series of poly(norbornene)-based ACIs were synthesized and deployed in hydrogen evolving AEMEL cathode electrodes. A balance between ionic conductivity, WU and ionomer swelling was achieved in the ACI by varying the IEC and degree of polymer cross-linking. It was found that higher IEC ACIs with light crosslinking are preferred in the HER electrode. Furthermore, such a configuration fine-tuned the WU and ionomer swelling to achieve optimum cell performance and reduce cell operating voltages.

08 HYDROGEN↗
DOE OSTI2020

Parameter Optimization Toolbox for NS-3 network optimization, NS-3 Parameter Optimization Framework [SWR-18-60]

This simulation-based parameter optimization framework is proposed to tune parameters of different types of communication networks using ns-3 to achieve the optimal network performance. It consists of three main components: an ns-3 packet reporting module; a sampler running simulations with all possible parameter sets for the input parameter variables by using a parallel executor at each generation; and a hybrid optimization algorithm for tuning configurable parameters of hybrid designs and application parameter variables. The proposed hybrid metaheuristic optimization algorithm combines an evolutionary algorithm with a gradient descent function to quickly achieve an approximate globally optimum solution. This software is designed to be used in a multi-core processing Linux environment and run over a long duration of time. The execution time varies depending mainly upon the nature of the ns-3 configuration being simulated. This software includes a custom ns-3 QoS measurement application which must be included with the ns-3 source code during installation of the software.

Hasandka, Adarsh↗
DOE OSTI2021

Bulb-style Kaplan Turbine - Run-of-the-river (ror) Hydro Governor And Turbine Model

This software package includes hydro governor and turbine models developed in Simulink (Mathworks Inc.) and RSCAD (RTDS Technologies) for low-head bulb-style Kaplan turbine hydropower plants. The model developed in Simulink has been designed and tuned to match the governor-turbine response for a real-world hydropower unit owned by Idaho Falls Power (IFP). The Simulink and RSCAD models enable real-time testing in a hardware-in-the-loop (HIL) implementation using OPAL-RT and RTDS digital real-time simulators, respectively. An automatic initialization for dynamic simulation has also been integrated to each model in both Simulink and RSCAD. The RSCAD initialization involves a novel python-based interfacing for automatic modification of the initial conditions in the model. The parameters of these models can be tuned to match other hydropower plants in this class.

Alam, SMShafiul↗
DOE OSTI2022

Tsuchinoko v1.0.0

Tsuchinoko is a Qt application for adaptive experiment execution and tuning. Live visualizations show details of measurements, and provide feedback on the adaptive engine's decision-making process. The parameters of the adaptive engine can also be tuned live to explore and optimize the search procedure. While Tsuchinoko is designed to allow custom adaptive engines to drive experiments, the gpCAM engine is a featured inclusion. This tool is based on a flexible and powerful Gaussian process regression at the core. A Tsuchinoko system includes 4 distinct components: the GUI client, an adaptive engine, and execution engine, and a core service. These components are separable to allow flexibility with a variety of distributed designs.

Pandolfi, Ronald↗
DOE OSTI2026

Janus v1.0

Janus provides a software framework for lightweight container management and orchestration. It's primary use cases are around deploying containerized services for high-performance data movement needs. Thus, Janus differentiates itself from systems like Kubernetes by tailoring the deployment of containers around network, storage, and host tuning optimizations. Janus uses the concept of profiles to capture repeatable deployment patterns and applies them to container execution across one or more endpoints. A Janus Agent component provides remote host tuning and monitoring capabilities.

Essiari, Abdelilah [Lawrence Berkeley National Lab↗
DOE OSTI2026

HydraGNN_GFM_FineTuning4Materials v1.0

This repository enables fine-tuning of the HydraGNN Predictive GFM 2026 — an open-source ensemble of pre-trained graph foundation models for atomistic materials modeling, developed at Oak Ridge National Laboratory. The GFM 2026 is freely available and downloadable via Globus from the OLCF Data Constellation (DOI: 10.13139/OLCF/2562660). Starting from these pre-trained weights, this repository provides a complete transfer learning pipeline for adapting the GFM ensemble to domain-specific molecular and materials property prediction tasks. It includes: 1) Utilities for ensemble fine-tuning with task-specific output heads 2) Example pipelines for eight widely-used materials and molecular datasets 3) Tools for model adaptation and head configuration 4) Data preprocessing utilities for each supported dataset 5) Benchmarking and evaluation scripts

Ungerboeck, Linda↗
DOE OSTI2021

Energy Exascale Earth System Model v2.0

First release of version 2 of the Energy Exascale Earth System Model. The atmosphere component remains EAM. Major changes since version 1 include: all column-physics parameterizations are computed on a separate grid that has approximately half the number of points of the dynamics grid, a new nonhydrostatic dynamical core (running in hydrostatic mode) with semi-Lagrangian tracer transport, CLUBB updated from v1 to v2, a new convective trigger (dCAPE/ULL) based on the dynamic Convective Available Potential Energy (CAPE) (dCAPE) and the Unrestricted Launch Level (ULL) concepts is used in ZM. minimum cloud droplet number changed, gravity wave drag energy conservation fixed and new tunings used, dust emission size distribution changed to emit more coarse dust particles The land component is still ELM. Major changes since version 1 include: using SNICAR-AD for radiation in snow to match the sea-ice model and fixing bugs in snow compaction and water state calculation. The ocean component remains MPAS-ocean. Major change since version 1 include: Redi isopycnal mixing has been updated, tested, and tuned in combination with the Gent-McWilliams parameterization, a sign error was fixed in the 3rd-order flux routines, the mesh used in low-resolution coupled cases was modified and the time steps adjusted, new regionally refined meshes were created, one focused on North America and another on the Southern Ocean, including ice shelf cavities. The sea-ice component remains MPAS-seaice. Major changes since version 1 include: A new heat- and freshwater-conserving coupling of frazil ice, turning off of SSH filtering, addition of SNICAR-AD and snow grain aging. The land-ice component remains MPAS-Albany-landIce (MALI) and is a static ice sheet. There is more out-of-the-box support for cryosphere configurations including the new regionally refined configuration around Antarctica with ice shelf cavities. The river model is MOSART. The half-degree river mesh was redone so it no longer treats Black and Caspian seas as ocean. The coupler remains cpl7/MCT. Major changes since version 1 include handling of ice shelf melt fluxes (heat / freshwater exchange with the ocean), and data icebergs. All components allow regional refinement of their meshes and two separate example of refinement, one in and around North America and one around Antarctica and the Southern Ocean, are provided.

E3SM Project, DOE↗
DOE OSTI2025

Analyzing and Exploring Training Recipes for Large-Scale Transformer-Based Weather Prediction

Abstract The rapid rise of deep learning (DL) in numerical weather prediction (NWP) has led to a proliferation of models which forecast atmospheric variables with comparable or superior skill than traditional physics-based NWP. However, among these leading DL models, there is a wide variance in both the training settings and architecture used. Further, the lack of thorough ablation studies makes it hard to discern which components are most critical to success. In this work, we show that it is possible to attain high forecast skill even with relatively off-the-shelf architectures, simple training procedures, and moderate compute budgets. Specifically, we train a minimally modified Swin Transformer V2 (SwinV2) on ERA5 data and find that it attains superior skill in terms of mean-square errors of deterministic forecasts when compared against the European Centre for Medium-Range Weather Forecasts’ Integrated Forecasting System (IFS). Almost all DL–NWP systems share a core set of hyperparameters and design decisions. To aid and expedite future DL–NWP research, we present an in-depth, systematic exploration of different loss functions, model sizes and depths, patch sizes, and multistep training objectives. We also examine the model performance with metrics beyond the typical accuracy (ACC) and RMSE and investigate how the performance scales with model size. Through our open-source code, scoring pipelines, and models, we share our findings on key aspects of the training pipeline. These ablations reduce the necessity for expensive hyperparameter tuning and lower the barrier to entry for future DL–NWP research. Significance Statement This study investigates the potential of using large-scale transformer-based models for weather prediction, showing that it is possible to achieve high forecast accuracy with simpler, off-the-shelf architectures. By training a minimally modified SwinV2 transformer on ERA5 data, we show that the model achieves competitive forecast skill in terms of mean-square error for key variables, outperforming the European Centre for Medium-Range Weather Forecasts’ Integrated Forecasting System (IFS) at all lead times. Our findings suggest that effective training strategies, such as multistep fine-tuning and channel-weighted losses, significantly enhance the model’s performance. However, we also highlight that these improvements come with trade-offs in other areas, such as ensemble spread and high-frequency spatial detail. This work highlights the promise of deep learning in improving weather forecasts, which could lead to better preparedness and response to weather events, ultimately benefiting society by providing more reliable weather predictions.

Willard, Jared D. [Lawrence Berkeley National Labo↗
DOE OSTI2021

Evaluation of KDP Estimation Algorithm Performance in Rain Using a Known-Truth Framework

Accurate estimation of specific differential phase ( K DP ) is necessary for rain rate estimation, attenuation correction, and hydrometeor classification algorithms. There are numerous published methods to process polarimetric radar observations of propagation differential phase shift (Φ DP ) and estimate K DP , but the corresponding K DP estimate uncertainty is unquantified. This study provides guidance on how commonly used K DP estimation algorithms perform in various environments. Here, we create numerous synthetic (“true”) K DP profiles, integrate over them to obtain “smoothed” Φ DP , and then add noise typical of S-band operational weather radar measurements. Each algorithm is applied to our noisy Φ DP profiles and compared to the true K DP profile such that the errors and uncertainty are quantified. The synthetic K DP profiles are Gaussian in shape, which allows systematic variations in their magnitude and width to determine how each algorithm performs in smooth, slowly changing K DP profiles, as well as steep profiles. Results demonstrate that algorithm performance is dependent on the Φ DP field received. These results are further supported by an error analysis of each algorithm for two more complicated synthetic K DP profiles. Some K DP algorithms allow users to change various tuning parameters; a subset of these tuning parameters is tested to provide guidance on how changing these parameters impacts algorithm performance. We then provide evidence that our known-truth framework provides insight into algorithm performance in observed data through two case studies.

54 ENVIRONMENTAL SCIENCES↗
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 OSTI2021

Mapping the combustion modes of a dual-fuel compression ignition engine

Compression-ignition (CI) engines can produce higher thermal efficiency (TE) and thus lower carbon dioxide (CO 2 ) emissions than spark-ignition (SI) engines. Unfortunately, the overall fuel economy of CI engine vehicles is limited by their emissions of nitrogen oxides (NO x ) and soot, which must be mitigated with costly, resource- and energy-intensive aftertreatment. NO x and soot could also be mitigated by adding premixed gasoline to complement the conventional, non-premixed direct injection (DI) of diesel fuel in CI engines. Several such “dual-fuel” combustion modes have been introduced in recent years, but these modes are usually studied individually at discrete conditions. This paper introduces a mapping system for dual-fuel CI modes that links together several previously studied modes across a continuous two-dimensional diagram. This system includes the conventional diesel combustion (CDC) and conventional dual-fuel (CDF) modes; the well-explored advanced combustion modes of HCCI, RCCI, PCCI, and PPCI; and a previously discovered but relatively unexplored combustion mode that is herein titled “Piston-split Dual-Fuel Combustion” or PDFC. Tests show that dual-fuel CI engines can simultaneously increase TE and lower NO x and/or soot emissions at high loads through the use of Partial HCCI (PHCCI). At low loads, PHCCI is not possible, but either PDFC or RCCI can be used to further improve NO x and/or soot emissions, albeit at slightly lower TE. These results lead to a “partial dual-fuel” multi-mode strategy of PHCCI at high loads and CDC at low loads, linked together by PDFC. Drive cycle simulations show that this strategy, when tuned to balance NO x and soot reductions, can reduce engine-out CO 2 emissions by about 1% while reducing NO x and soot by about 20% each with respect to CDC. This increases emissions of unburnt hydrocarbons (UHC), still in a treatable range (2.0 g/kWh) but five times as high as CDC, requiring changes in aftertreatment strategy.

Engineering↗
DOE OSTI2024

Clostridium autoethanogenum alters cofactor synthesis, redox metabolism, and lysine-acetylation in response to elevated H 2 :CO feedstock ratios for enhancing carbon capture efficiency

Clostridium autoethanogenum is an acetogenic bacterium that autotrophically converts carbon monoxide (CO) and carbon dioxide (CO 2 ) gases into bioproducts and fuels via the Wood–Ljungdahl pathway (WLP). To facilitate overall carbon capture efficiency, the reaction stoichiometry requires supplementation of hydrogen at an increased ratio of H 2 :CO to maximize CO 2 utilization; however, the molecular details and thus the ability to understand the mechanism of this supplementation are largely unknown. In order to elucidate the microbial physiology and fermentation where at least 75% of the carbon in ethanol comes from CO 2 , we established controlled chemostats that facilitated a novel and high (11:1) H 2 :CO uptake ratio. We compared and contrasted proteomic and metabolomics profiles to replicate continuous stirred tank reactors (CSTRs) at the same growth rate from a lower (5:1) H 2 :CO condition where ~ 50% of the carbon in ethanol is derived from CO 2 . Our hypothesis was that major changes would be observed in the hydrogenases and/or redox-related proteins and the WLP to compensate for the elevated hydrogen feed gas. Our analyses did reveal protein abundance differences between the two conditions largely related to reduction–oxidation (redox) pathways and cofactor biosynthesis, but the changes were more minor than we would have expected. While the Wood–Ljungdahl pathway proteins remained consistent across the conditions, other post-translational regulatory processes, such as lysine-acetylation, were observed and appeared to be more important for fine-tuning this carbon metabolism pathway. Metabolomic analyses showed that the increase in H 2 :CO ratio drives the organism to higher carbon dioxide utilization resulting in lower carbon storages and accumulated fatty acid metabolite levels. This research delves into the intricate dynamics of carbon fixation in C. autoethanogenum, examining the influence of highly elevated H 2 :CO ratios on metabolic processes and product outcomes. The study underscores the significance of optimizing gas feed composition for enhanced industrial efficiency, shedding light on potential mechanisms, such as post-translational modifications (PTMs), to fine-tune enzymatic activities and improve desired product yields.

09 BIOMASS FUELS↗
DOE OSTI2026

When does global attention help: a unified empirical study on atomistic graph learning

Graph neural networks (GNNs) are widely used as surrogates for costly experiments and first-principles simulations to study the behavior of compounds at atomistic scale, and their architectural complexity is constantly increasing to enable the modeling of complex physics. While most recent GNNs combine more traditional message passing neural networks (MPNNs) layers to model short-range interactions with more advanced graph transformers (GTs) with global attention mechanisms to model long-range interactions, it is still unclear when global attention mechanisms provide real benefits over well-tuned MPNN layers due to inconsistent implementations, features, or hyperparameter tuning. We introduce the first unified, reproducible benchmarking framework–built on HydraGNN–that enables seamless switching among four controlled model classes: MPNN, MPNN with chemistry/topology encoders, GPS-style hybrids of MPNN with global attention, and fully fused localglobal models with encoders. Using seven diverse open-source datasets for benchmarking across regression and classification tasks, we systematically isolate the contributions of message passing, global attention, and encoder-based feature augmentation. Our study shows that encoder-augmented MPNNs form a robust baseline, while fused localglobal models yield the clearest benefits for properties governed by long-range interaction effects. We further quantify the accuracycompute trade-offs of attention, reporting its overhead in memory. Together, these results establish the first controlled evaluation of global attention in atomistic graph learning and provide a reproducible testbed for future model development.

Equivariant graph neural networks↗
DOE OSTI2020

In-situ measurement of pyrolysis and combustion gases from biomass burning using swept wavelength external cavity quantum cascade lasers

Broadband high-speed absorption spectroscopy using swept-wavelength external cavity quantum cascade lasers (ECQCLs) is applied to measure multiple pyrolysis and combustion gases in biomass burning experiments. Two broadly-tunable swept-ECQCL systems were used, with the first tuned over a range of 2089-2262 cm -1 (4.42 – 4.79 µm) to measure spectra of CO 2 , H 2 O, and CO. The second was tuned over a range of 920-1150 cm -1 (8.70 – 10.9 µm) to measure spectra of ammonia (NH 3 ), ethene (C 2 H 4 ), and methanol (MeOH). Absorption spectra were measured continuously at a 100 Hz rate throughout the burn process, including inhomogeneous flame regions, and analyzed to determine time-resolved gas concentrations and temperature. The results provide in-situ, dynamic information regarding gas-phase species as they are generated, close to the biomass fuel source.

09 BIOMASS FUELS↗
DOE OSTI2020

Shallow-etched thin-film lithium niobate waveguides for highly-efficient second-harmonic generation

High-fidelity periodic poling over long lengths is required for robust, quasi-phase-matched second-harmonic generation using the fundamental, quasi-TE polarized waveguide modes in a thin-film lithium niobate (TFLN) waveguide. Here, a shallow-etched ridge waveguide is fabricated in x-cut magnesium oxide doped TFLN and is poled accurately over 5 mm. The high fidelity of the poling is demonstrated over long lengths using a non-destructive technique of confocal scanning second-harmonic microscopy. We report a second-harmonic conversion efficiency of up to 939 %.W −1 (length-normalized conversion efficiency 3757 %.W −1 .cm −2 ), measured at telecommunications wavelengths. The device demonstrates a narrow spectral linewidth (1 nm) and can be tuned precisely with a tuning characteristic of 0.1 nm/°C, over at least 40 °C without measurable loss of efficiency.

Zhao, Jie (ORCID:0000000306230125)↗
DOE OSTI2020

Plasmonic nanoarcs: a versatile platform with tunable localized surface plasmon resonances in octave intervals

The tunability of the longitudinal localized surface plasmon resonances (LSPRs) of metallic nanoarcs is demonstrated with key relationships identified between geometric parameters of the arcs and their resonances in the infrared. The wavelength of the LSPRs is tuned by the mid-arc length of the nanoarc. The ratio between the attenuation of the fundamental and second order LSPRs is governed by the nanoarc central angle. Beneficial for plasmonic enhancement of harmonic generation, these two resonances can be tuned independently to obtain octave intervals through the design of a non-uniform arc-width profile. Because the character of the fundamental LSPR mode in nanoarcs combines an electric and a magnetic dipole, plasmonic nanoarcs with tunable resonances can serve as versatile building blocks for chiroptical and nonlinear optical devices.

42 ENGINEERING↗
DOE OSTI2020

Effect of heating/cooling dynamics in the hysteresis loop and tunable IR emissivity of VO 2 thin films

We experimentally investigate the semiconductor-to-metal transition (SMT) in vanadium dioxide thin films using an infrared thermographic technique. During the semiconductor to metal phase change process, VO 2 optical properties dynamically change and infrared emission undergoes a hysteresis loop due to differences between heating and cooling stages. The shape of the hysteresis loop was accurately monitored under different dynamic heating/cooling rates. In order to quantify and understand the effects of different rates, we used a numerical modelling approach in which a VO 2 thin layer was modeled as metamaterial. The main experimental findings are interpreted assuming that both the rate of formation and shape of metallic inclusions are tuned with the heating/cooling rate. The structural transition from monoclinic to tetragonal phases is the main mechanism for controlling the global properties of the phase transition. However, our experimental results reveal that the dynamics of the heating/cooling process can become a useful parameter for further tuning options and lays out a macroscopic optical sensing scheme for the microscopic phase change dynamics of VO 2 . Our study sheds light on phase-transition dynamics and their effect on the infrared emission spectra of VO 2 thin films, therefore enabling the heating/cooling rate to be an additional parameter to control infrared emission characteristics of thermal emitters. The hysteresis loop represents the phase coexistence region, thus being of fundamental importance for several applications, such as the operation of radiative thermal logic elements based on phase transition materials. For such applications, the phase transition region is shifted for heating and cooling processes. We also show that, depending on the way the phase change elements are heated, the temperature operation range will be slightly modified.

42 ENGINEERING↗
DOE OSTI2024

Arbitrary electro-optic bandwidth and frequency control in lithium niobate optical resonators

In situ tunable photonic filters and memories are important for emerging quantum and classical optics technologies. However, most photonic devices have fixed resonances and bandwidths determined at the time of fabrication. Here we present an in situ tunable optical resonator on thin-film lithium niobate. By leveraging the linear electro-optic effect, we demonstrate widely tunable control over resonator frequency and bandwidth on two different devices. We observe up to ∼50 × tuning in the bandwidth over ∼50 V with linear frequency control of ∼230 MHz/V. We also develop a closed-form model predicting the tuning behavior of the device. This paves the way for rapid phase and amplitude control over light transmitted through our device.

42 ENGINEERING↗