Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Hybrid model”

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 109 records · Page 6
All sourcesNASA NTRSDOE OSTI
NASA NTRS2006

Estimating Resolution Lengths of Hybrid Turbulence Models

A two-stage procedure has been devised for estimating the spatial resolution achievable in the simulation of a given flow on a given computational grid by a computational fluid dynamics (CFD) code that incorporates a hybrid model of turbulence. The hybrid models to which this procedure is especially relevant are those of the Reynolds-averaged Navier-Stokes (RANS) and the partial-averaged Navier-Stokes (PANS) approaches. This procedure represents the first step toward adding variable-resolution turbulence-modeling capabilities to CFD codes as part of a continuing effort to increase the accuracy and robustness of CFD simulations of unsteady flows. Some background information is prerequisite to a meaningful summary of the procedure. Among experts in CFD, it is well known that combination of the Reynolds-averaged Navier-Stokes (RANS) approach and eddy-viscosity turbulence models offers limited capability for simulating unsteady and complex flows. The RANS approach includes an assumption that most of the energy in a given flow is modeled through turbulence-transport equations and is resolved in a computational grid used to simulate the flow. RANS also overpredicts eddy viscosity, thereby yielding excessive damping of unsteady motion. The eddy viscosity attains an unphysically large value because of unresolved scales, and suppresses most temporal and spatial fluctuations in the resolved flow field. One approach used to overcome this deficiency is to provide a mechanism for the RANS equations to resolve motion only on the largest scales and to use a hybrid model to represent effects at smaller scales. The RANS approach involves the use of a standard two-equation turbulence model in which the effect of turbulence is summarized by a viscosity that is a function of (1) the time-averaged kinetic- energy density (k) associated with the local fluctuating (turbulent) component of flow and (2) the time-averaged rate of dissipation of the turbulent-kinetic- energy density ( ). In PANS, which was developed to overcome the grid dependency associated with other hybrid turbulence models (including that of RANS), the standard two-equation turbulence model is replaced by another two-equation model in which one solves for the previously unresolved k and , which are now allowed to vary in space and time.

Abdol-Hamid, Khaled S.↗
DOE OSTI2026

Deployment of Traditional and Hybrid Machine Learning for Critical Heat Flux Prediction in the CTF Thermal-Hydraulics Code

Critical heat flux (CHF) marks the transition from nucleate to film boiling, where heat transfer to the working fluid can rapidly deteriorate. Accurate CHF prediction is essential for efficiency, safety, and preventing equipment damage, particularly in nuclear reactors. Although widely used, empirical correlations frequently exhibit discrepancies when compared to experimental data, limiting their reliability in diverse operational conditions. Traditional machine learning (ML) approaches have demonstrated potential for CHF prediction but often suffer from limited interpretability, data scarcity, and insufficient knowledge of physical principles. Hybrid model approaches, which combine data-driven ML with base models, mitigate these concerns by incorporating prior knowledge of the domain. This study integrates an externally trained purely data-driven ML model and two hybrid models (using the Biasi and Bowring CHF correlations) within the CTF subchannel code via a custom Fortran framework. Performance was evaluated using two validation cases: a subset of the Nuclear Regulatory Commission (NRC) CHF database and the Bennett dryout experiments. In both cases, the hybrid models demonstrated significantly lower error metrics compared to conventional empirical correlations, with the best models often reducing relative error by about 5 percentage points. The pure ML model achieved comparable accuracy, outperforming the hybrid Biasi model in the NRC test case (3.3% versus 5.5% relative error) but exhibiting slightly higher error against the hybrid Bowring model in the Bennett test case (7.7% versus 6.1%). Trend analysis of error parity indicated that ML-based models reduced the tendency for CHF overprediction, improving overall accuracy. These results demonstrate that ML-based CHF models can be effectively integrated into subchannel codes and could potentially increase performance compared to conventional methods.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2026

Trigonometric continuous-variable gates and hybrid quantum simulations of the sine-Gordon model

Hybrid qubit-qumode quantum computing platforms provide a natural setting for simulating interacting bosonic quantum field theories. However, existing continuous-variable gate constructions rely predominantly on polynomial functions of canonical quadratures. In this work, we introduce a complementary universality paradigm based on trigonometric continuous-variable gates, which enable a Fourier-like representation of bosonic operators and are particularly well suited for periodic and non-perturbative interactions. We present an ancilla-based framework for implementing trigonometric gates with arguments given by arbitrary Hermitian functions of qumode quadratures. The protocol yields unitary gates deterministically, and non-unitary gates through probabilistic post-selection. As a concrete application, we develop a hybrid qubit-qumode quantum simulation of the lattice sine-Gordon model. Using these gates, we prepare ground states via quantum imaginary-time evolution, simulate real-time dynamics, compute time-dependent vertex two-point correlation functions, and extract quantum kink profiles under topological boundary conditions. Our results demonstrate that trigonometric continuous-variable gates provide a physically natural framework for simulating interacting field theories on near-term hybrid quantum hardware, while establishing a parallel route to universality beyond polynomial gate constructions. We expect that the trigonometric gates introduced here to find broader applications, including quantum simulations of condensed matter systems, quantum chemistry, and biological models.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
NASA NTRS2014

Model-Invariant Hybrid LES-RANS Computation of Separated Flow Past Periodic Hills

The requirement that physical quantities not vary with a hybrid LESRANS model's blending parameter imposes conditions on the computation that lead to better results across LES-RANS transitions. This promises to allow placement of those transitions so that LES is performed only where required by the physics, improving computational efficiency. The approach is applied to separated flow past periodic hills, where good predictions of separation-bubble size are seen due to the gradual, controlled, LES-RANS transition and the resulting enhanced near-wall eddy viscosity.

Woodruff, Stephen↗
NASA NTRS2016

Heating and Acceleration of Solar Wind Ions by Turbulent Wave Spectrum in Inhomogeneous Expanding Plasma

Near the Sun (< 10Rs) the acceleration, heating, and propagation of the solar wind are likely affected by the background inhomogeneities of the magnetized plasma. The heating and the acceleration of the solar wind ions by turbulent wave spectrum in inhomogeneous plasma is studied using a 2.5D hybrid model. The hybrid model describes the kinetics of the ions, while the electrons are modeled as massless neutralizing fluid in an expanding box approach. Turbulent magnetic fluctuations dominated by power-law frequency spectra, which are evident from in-situ as well as remote sensing measurements, are used in our models. The effects of background density inhomogeneity across the magnetic field on the resonant ion heating are studied. The effect of super- Alfvenic ion drift on the ion heating is investigated. It is found that the turbulent wave spectrum of initially parallel propagating waves cascades to oblique modes, and leads to enhanced resonant ion heating due to the inhomogeneity. The acceleration of the solar wind ions is achieved by the parametric instability of large amplitude waves in the spectrum, and is also affected by the inhomogeneity. The results of the study provide the ion temperature anisotropy and drift velocity temporal evolution due to relaxation of the instability. The non-Maxwellian velocity distribution functions (VDFs) of the ions are modeled in the inhomogeneous solar wind plasma in the acceleration region close to the Sun.

Ofman, Leon↗
DOE OSTI2021

Multifidelity computing for coupling full and reduced order models

Hybrid physics-machine learning models are increasingly being used in simulations of transport processes. Many complex multiphysics systems relevant to scientific and engineering applications include multiple spatiotemporal scales and comprise a multifidelity problem sharing an interface between various formulations or heterogeneous computational entities. To this end, we present a robust hybrid analysis and modeling approach combining a physics-based full order model (FOM) and a data-driven reduced order model (ROM) to form the building blocks of an integrated approach among mixed fidelity descriptions toward predictive digital twin technologies. At the interface, we introduce a long short-term memory network to bridge these high and low-fidelity models in various forms of interfacial error correction or prolongation. The proposed interface learning approaches are tested as a new way to address ROM-FOM coupling problems solving nonlinear advection-diffusion flow situations with a bifidelity setup that captures the essence of a broad class of transport processes.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRS1985

A study of Reynolds-stress closure model

A hybrid model of the Reynolds stress closure was developed. This model was tested for various sizes of step flow, and the computed Reynolds stress behavior was compared with experimental data. The third order closure model was reviewed. Transport equations for the triple velocity correlation were developed and implemented in a numerical code to evaluate the behavior of the triple velocity products in various regions of the flow field including recirculating, reattaching, and redeveloping flow domains.

Amano, R. S.↗
NASA NTRS2014

Titan's Plasma Environment : 3D Hybrid Kinetic Modeling of the TA Flyby and Comparison with CAPS-ELS and RPWS LP Observations

In this report we discuss the global plasma environment of the TA flyby from the perspective of 3D hybrid modeling. In our model the background, pickup, and ionospheric ions are considered as particles, whereas the electrons are described as a fluid. In homogeneous photoionization, electron-impact ionization and charge exchange are included in our model. We also take into account the collisions between the ions and neutrals. Our modeling shows that mass loading of the background plasma (H(+), O(+)) by pick up ions H2(+), CH4(+) and N2(+) differs from theT9 encounter simulations when O(+) ions are not introduced into the background plasma. In our hybrid-modeling we use Chamberlain profiles for the atmospheric components. We also include a simple ionosphere model with average mass M=28 amu ions that were generated inside the ionosphere. Titan's interior is considered as a weakly conducting body. Special attention has been paid to comparing the simulated pickup ion density distribution with CAPS-ELS and with RPWS LP observations by the Cassini-Huygens spacecraft along theta trajectory. Our modeling shows an asymmetry of the ion density distribution and the magnetic field, including the formation of Alfven wing-like structures.

induced magnetospheres↗
NASA NTRS2016

Global Effects of Transmitted Shock Wave Propagation Through the Earth's Inner Magnetosphere: First Results from 3-D Hybrid Kinetic Modeling

We use a new hybrid kinetic model to simulate the response of ring current, outer radiation belt, and plasmaspheric particle populations to impulsive interplanetary shocks. Since particle distributions attending the interplanetary shock waves and in the ring current and radiation belts are non-Maxwellian, waveparticle interactions play a crucial role in energy transport within the inner magnetosphere. Finite gyroradius effects become important in mass loading the shock waves with the background plasma in the presence of higher energy ring current and radiation belt ions and electrons. Initial results show that shocks cause strong deformations in the global structure of the ring current, radiation belt, and plasmasphere. The ion velocity distribution functions at the shock front, in the ring current, and in the radiation belt help us determine energy transport through the Earth's inner magnetosphere.

Lipatov, A. S.↗
NASA NTRS2020

Concerning the Interaction of A Transmitted Interplanetary Impulse With A Plasmaspheric Drainage Plume: First Results From 3-D Hybrid Kinetic Modeling

We present a new hybrid kinetic model to simulate the response of plasmaspheric drainage plumes to impulsive interplanetary pressure pulses. Since particle distributions attending the interplanetary pulses and in the drainage plume are non-Maxwellian, wave-particle interactions play a crucial role in energy transport within and outside the plumes. Finite gyroradius effects become important in mass loading of the transmitted impulse with the drainage plume ions. A forward-reverse shock structure develops from the initial step-like transmitted shock. First results show that the impulse causes strong deformations in the global structure of the plume. The anisotropic ion velocity distribution functions at the impulse front and inside the plume help us determine energy transport via wave-particle interactions throughout the Earth’s inner magnetosphere.

A S Lipatov↗
NASA NTRSDate not supplied

An Active Model Split for Hybrid RANS/LES of Compressible Flows

Hybrid RANS/LES models have consistently shown superior accuracy over RANS in predictions of massively separated flows. However, these models have experienced difficulties with modeled stress depletion, smooth-wall separated flow, and shock-separated flows. The Active Model Split (AMS) hybridization was created to address some of the fundamental shortcomings of traditional hybrid RANS/LES models. While previous work considered incompressible flows, this paper demonstrates how the AMS framework can be extended to address compressible flows. AMS-based hybrid models for the turbulent heat flux and turbulent transport terms in the total energy equation are presented. This compressible hybrid RANS/LES framework is demonstrated on a subsonic channel, a supersonic channel, and a transonic shock-boundary layer interaction. Record includes video of presentation, to view download the mp4 video attached run time of 10 mins 18 secs.

hybrid RANS/LES↗
DOE OSTI2026

Hybrid Dynamic Modeling of Smart Inverter

This letter proposes a novel hybrid method for assessing grid-connected three-phase converter interfaced resources (CIR) dynamics with the IEEE standard 1547-2018 grid support functions (GSFs), which blends physics and data-driven techniques. First, the letter derives an analytical model of a CIR to represent the internal physics and data-driven model (DDM) using a system identification algorithm to represent the rest of the dynamics, including the GSF. The derived hybrid model combines the analytical model of CIR and DDM, which balances accuracy and flexibility and is compared with the detailed switched model. Furthermore, the efficacy of the proposed approach to represent the advanced CIR dynamics is substantiated by power hardware-in-the-loop experiment data where real measurements from a commercial CIR are used to cross-validate the proposed approach. Furthermore, the results indicate that despite simple, the hybrid model accurately reproduces the dynamics of the detailed CIR model with an acceptable accuracy.

Data-driven model↗
NASA NTRS1993

Hybrid experimental/analytical models of structural dynamics - Creation and use for predictions

An original complete methodology for the construction of predictive models of damped structural vibrations is introduced. A consistent definition of normal and complex modes is given which leads to an original method to accurately identify non-proportionally damped normal mode models. A new method to create predictive hybrid experimental/analytical models of damped structures is introduced, and the ability of hybrid models to predict the response to system configuration changes is discussed. Finally a critical review of the overall methodology is made by application to the case of the MIT/SERC interferometer testbed.

Balmes, Etienne↗
DOE OSTI2026

Ripening of Rh Nanoparticle Catalysts in Reverse Water–Gas Shift via a Data-Driven Model Combining Physics, Theory, and Experiment

Degradation via sintering is an ongoing challenge that impedes the broad commercial success of supported metallic nanoparticle catalysts. To mitigate degradation via informed catalyst design and process operations, here we aim to disambiguate the underlying mechanisms of sintering by combining theory and experiment in a quantitative framework. While mechanistic sintering models exist, they only model a single sintering pathway, even though multiple sintering mechanisms can occur simultaneously or dominate at different stages of the process. Data-driven machine learning models have emerged as a means to represent complex processes through data regression. However, machine learning models have very large data needs and lack mechanistic insights due to their black-box encoding. To develop an interpretive model of catalyst degradation via sintering, we constructed a hybrid model combining mechanistic “physics-based” models and data-driven methods to obtain both reliable predictions and mechanistic insights regarding experimentally observed sintering phenomena. Focusing on nanoparticle sintering in the Rh–TiO 2 catalyst for the reverse water–gas shift (RWGS) reaction, the hybrid model couples a mechanistic term for Ostwald ripening with energy values calculated via density functional theory (DFT) with a parametric, data-driven discrepancy function term for unmodeled mechanisms. The hybrid model is trained using Bayesian inference with data collected from small-angle X-ray scattering (SAXS) in situ experiments wherein average nanoparticle diameter versus time was measured at three relevant operating temperatures. The calibrated hybrid model results show that an Ostwald ripening-only model parameterized with fixed DFT energies does not fully capture the time and temperature dependence of the SAXS-observed sintering kinetics, and that an additional functional contribution, or DFT energy calibration, is required to reconcile simulation and experiment. Analysis of the hybrid-model error confirms that the hybrid model outperforms both the purely mechanistic and purely data-driven alternatives in terms of expected predictive accuracy for time-evolving average particle sizes. Furthermore, the results support the hypothesis that the Ostwald ripening mechanism is less important for explaining the sintering phenomena as operating temperature increases under an assumed fixed DFT parameterization. This could be explained in one of two ways: either latent, unmodeled sintering mechanisms dominate at higher temperatures, or the DFT uncertainty increases with temperature. The proposed modeling approach directly links theory to experiments and simulations via a statistical hybrid modeling framework and can be extended to other catalytic systems to improve predictive models and mechanistic understanding.

Bayesian hybrid modeling↗
DOE OSTI2026

Hybrid combustion modeling approach for turbulent jet ignition in natural-gas pre-chamber spark-ignition engines at high EGR

Here, this study presented a hybrid modeling approach for simulating turbulent jet ignition and combustion processes in a natural-gas pre-chamber spark-ignition engine operating under exhaust gas recirculation (EGR) diluted conditions. In-depth analyses of experimental data and simulation results from previous work [Chinnathambi et al., ICEF2021-67836; Kim et al., Fuel 409: 137815, 2026] revealed two key findings: (i) the magnitude of pressure difference between the pre-chamber and main chamber ($∆P_{PC-MC}$) was positively correlated with the combustion duration from the moment of $∆P_{PC-MC}=0$ to the point of 5% mass fraction burned, with larger $∆P_{PC-MC}$ associated with longer duration; and (ii) the turbulent combustion regime in the main chamber transitioned from the broken reaction zone to the corrugated flamelet regime, with the Karlovitz number exceeding 100 immediately after turbulent hot jets were ejected from nozzles, coinciding with observed local extinction events. To accurately simulate the entire combustion process, a hybrid approach was developed under Reynolds-Averaged Navier Stokes framework, combining the G-equation model for pre-chamber combustion with the multi-zone well-stirred reactor approach and a turbulence-chemistry interaction (TCI) submodel for main chamber combustion. The TCI submodel accounted for the attenuation of reaction rates due to turbulent strain and modeled local extinction by suppressing reaction rates under certain flow and flame conditions. When applied to three EGR rate conditions toward the dilution limit, the hybrid modeling approach accurately reproduced experimental data in terms of cylinder pressure, apparent heat release rate, and the observed positive correlation, including the delayed onset of main chamber combustion—a feature not captured by existing combustion models.

computational fluid dynamics simulation↗
DOE OSTI2023

Physics-guided neural networks with engineering domain knowledge for hybrid process modeling

As neural networks are more frequently used to solve problems in science and engineering, the methods used to incorporate scientific knowledge into these networks are becoming increasingly complex. Here, this work breaks down these complicated techniques into a set of basic strategies which can easily be applied to diverse situations. Several novel neural networks are built using the categories laid out in this work. These networks are tested on simulated data from a continuous stirred tank reactor (CSTR) model to evaluate the advantages provided by each network. The three points demonstrated in this work are: (1) architectural hybrid models can speed up convergence and reduce the amount of data necessary to train a model; (2) adding a physics-guided loss function can improve model generalization and make models more physically consistent; (3) using physics-guided initialization and transfer learning improves accuracy and speeds up convergence, but can harm generalizability if used incorrectly.

42 ENGINEERING↗
DOE OSTI2023

Status Report on Thermal Delivery Modeling in HYBRID

This report details the modeling efforts consistent with the HYBRID repository of the Integrated Energy Systems (IES) program, motivated to develop information data tables characterizing the thermodynamic losses associated with transporting thermal energy over large distances. The described model establishes a methodology to currently characterize the impacts of certain parameters such as distance, piping geometry, and heat transfer fluid among others as well as sets up ability to interconnect the delivery model into Modelica IES simulations in the future. This report describes the modeling efforts completed and describes some of the future studies that will be done using results produced by the modeling efforts.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
NASA NTRS2018

Effects of Multiscale Phase-Mixing and Interior Conductance in the Lunar-like Pickup Ion Plasma Wake. First Results from 3-D Hybrid Kinetic Modeling

The study of multiscale pickup ion phase-mixing in the lunar plasma wake with a hybrid model is the main subject of our investigation in this paper. Photoionization and charge exchange of protons with the lunar exosphere are the ionization processes included in our model. The computational model includes the self-consistent dynamics of the light (hydrogen ions or hydrogen molecule ion or helium ion), and heavy (sodium ion ) pickup ions. The electrons are considered as a fluid. The lunar interior is considered as a weakly conducting body. In this paper we considered for the first time the cumulative effect of heavy neutrals in the lunar exosphere (e.g., Aluminum, Argon), an effect which was simulated with one species of sodium ion but with a tenfold increase in total production rates. We find that various species produce various types of plasma tail in the lunar plasma wake. Specifically, sodium ion and helium ion pickup ions form a cycloid-like tail, whereas the hydrogen ion or hydrogen molecule ion pickup ions form a tail with a high density core and saw-like periodic structures in the flank region. The length of these structures varies from 1:5 R(sub M) to 3:3 R(sub M) depending on the value of gyro radius for hydrogen ion or hydrogen molecule ion pickup ions. The light pickup ions produce more symmetrical jump in the density and magnetic field at the Mach cone which is mainly controlled by the conductivity of the interior, an effect previously unappreciated. Although other pickup ion species had little effect on the nature of the interaction of the Moon with the solar wind, the global structure of the lunar tail in these simulations appeared quite different when the hydrogen molecule ion production rate was high.

Lipatov, A.S.↗