Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Manifold 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 145 records · Page 8
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

A Tailored Convolutional Neural Network for Nonlinear Manifold Learning of Computational Physics Data Using Unstructured Spatial Discretizations

In this work, we propose a nonlinear manifold learning technique based on deep convolutional autoencoders that is appropriate for model order reduction of physical systems in complex geometries. Convolutional neural networks have proven to be highly advantageous for compressing data arising from systems demonstrating a slow-decaying Kolmogorov n-width. However, these networks are restricted to data on structured meshes. Unstructured meshes are often required for performing analyses of real systems with complex geometry. Our custom graph convolution operators based on the available differential operators for a given spatial discretization effectively extend the application space of deep convolutional autoencoders to systems with arbitrarily complex geometry that are typically discretized using unstructured meshes. We propose sets of convolution operators based on the spatial derivative operators for the underlying spatial discretization, making the method particularly well suited to data arising from the solution of partial differential equations. We demonstrate the method using examples from heat transfer and fluid mechanics and show better than an order of magnitude improvement in accuracy over linear methods.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Improved surrogates in inertial confinement fusion with manifold and cycle consistencies

Neural networks have become the method of choice in surrogate modeling because of their ability to characterize arbitrary, high-dimensional functions in a data-driven fashion. This paper advocates for the training of surrogates that are 1) consistent with the physical manifold, resulting in physically meaningful predictions, and 2) cyclically consistent with a jointly trained inverse model; i.e., backmapping predictions through the inverse results in the original input parameters. We find that these two consistencies lead to surrogates that are superior in terms of predictive performance, are more resilient to sampling artifacts, and tend to be more data efficient. Using inertial confinement fusion (ICF) as a test-bed problem, we model a one-dimensional semianalytic numerical simulator and demonstrate the effectiveness of our approach.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
NASA NTRS1985

Duct flow nonuniformities study for space shuttle main engine

To improve the Space Shuttle Main Engine (SSME) design and for future use in the development of generation rocket engines, a combined experimental/analytical study was undertaken with the goals of first, establishing an experimental data base for the flow conditions in the SSME high pressure fuel turbopump (HPFTP) hot gas manifold (HGM) and, second, setting up a computer model of the SSME HGM flow field. Using the test data to verify the computer model it should be possible in the future to computationally scan contemplated advanced design configurations and limit costly testing to the most promising design. The effort of establishing and using the computer model is detailed. The comparison of computational results and experimental data observed clearly demonstrate that computational fluid mechanics (CFD) techniques can be used successfully to predict the gross features of three dimensional fluid flow through configurations as intricate as the SSME turbopump hot gas manifold.

Thoenes, J.↗
DOE OSTI2020

Efficient bifurcation and tabulation of multi-dimensional combustion manifolds using deep mixture of experts: An a priori study

This work describes and validates an approach for autonomously bifurcating turbulent combustion man-ifolds to divide regression tasks amongst specialized artificial neural networks (ANNs). This approach relies on the mixture of experts (MoE) framework, where each neural network is trained to be specialized in a given portion of the input space. The assignment of different input regions to the experts is determined by a gating network, which is a neural network classifier. In some previous studies, it has been demonstrated that bifurcation of a complex combustion manifold and fitting different ANNs for each part leads to better fits or faster inference speeds. However, the manner of bifurcation in these studies was based on heuristic approaches or clustering techniques. In contrast, the proposed technique enables automatic bifurcation using non-linear planes in high-dimensional turbulent combustion manifolds that are often associated with complex behavior due to different dominating physics in various zones. Furthermore, the proposed concept is validated using 4-dimensional (4D) and 5D flamelet tables, showing that the errors obtained with a given network size, or conversely the network size required to achieve a given accuracy, is considerably reduced. The effect of the number of experts on inference speed is also investigated, showing that by increasing the number of experts from 1 to 8, the inference time can be approximately reduced by a factor of two. Moreover, it is shown that the MoE approach divides the input manifold in a physically intuitive manner, suggesting that the MoE framework can elucidate high-dimensional datasets in a physically meaningful way.

42 ENGINEERING↗
DOE OSTI2022

Reduced order modeling for flow and transport problems with Barlow Twins self-supervised learning

Abstract We propose a unified data-driven reduced order model (ROM) that bridges the performance gap between linear and nonlinear manifold approaches. Deep learning ROM (DL-ROM) using deep-convolutional autoencoders (DC–AE) has been shown to capture nonlinear solution manifolds but fails to perform adequately when linear subspace approaches such as proper orthogonal decomposition (POD) would be optimal. Besides, most DL-ROM models rely on convolutional layers, which might limit its application to only a structured mesh. The proposed framework in this study relies on the combination of an autoencoder (AE) and Barlow Twins (BT) self-supervised learning, where BT maximizes the information content of the embedding with the latent space through a joint embedding architecture. Through a series of benchmark problems of natural convection in porous media, BT–AE performs better than the previous DL-ROM framework by providing comparable results to POD-based approaches for problems where the solution lies within a linear subspace as well as DL-ROM autoencoder-based techniques where the solution lies on a nonlinear manifold; consequently, bridges the gap between linear and nonlinear reduced manifolds. We illustrate that a proficient construction of the latent space is key to achieving these results, enabling us to map these latent spaces using regression models. The proposed framework achieves a relative error of 2% on average and 12% in the worst-case scenario (i.e., the training data is small, but the parameter space is large.). We also show that our framework provides a speed-up of $$7 \times 10^{6}$$ 7 × 10 6 times, in the best case, and $$7 \times 10^{3}$$ 7 × 10 3 times on average compared to a finite element solver. Furthermore, this BT–AE framework can operate on unstructured meshes, which provides flexibility in its application to standard numerical solvers, on-site measurements, experimental data, or a combination of these sources.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Machine learning for the identification of phase transitions in interacting agent-based systems: A Desai-Zwanzig example

Deriving closed-form analytical expressions for reduced-order models, and judiciously choosing the closures leading to them, has long been the strategy of choice for studying phase- and noise-induced transitions for agent-based models (ABMs). In this paper, we propose a data-driven framework that pinpoints phase transitions for an ABM—the Desai-Zwanzig model—in its mean-field limit, using a smaller number of variables than traditional closed-form models. To this end, we use the manifold learning algorithm Diffusion Maps to identify a parsimonious set of data-driven latent variables, and we show that they are in one-to-one correspondence with the expected theoretical order parameter of the ABM. We then utilize a deep learning framework to obtain a conformal reparametrization of the data-driven coordinates that facilitates, in our example, the identification of a single parameter-dependent ordinary differential equation (ODE) in these coordinates. Additionally, we identify this ODE through a residual neural network inspired by a numerical integration scheme (forward Euler). We then use the identified ODE—enabled through an odd symmetry transformation—to construct the bifurcation diagram exhibiting the phase transition.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Performance Characterization of a Natural Gas–Air Rotating Detonation Engine

An experimental study of a rotating detonation engine (RDE) operating with natural gas and air at elevated chamber pressures and air preheat temperatures was conducted to quantify its performance at conditions representative of land-based power generation gas turbine engines. Here, the thrust produced by the combustor was measured to characterize its work output potential. High-frequency pressure transducers and broadband chemiluminescence measurements of the flame provided information about the wave structure and dynamics. Analysis of common performance metrics demonstrated the necessity of normalizing any RDE performance parameter by the driving system potential, typically the reactant manifold pressure. Application of a thermodynamic performance model to a generic RDE identified the area ratio between the RDE exhaust and injection throats as the primary parameter affecting delivered pressure gain. The model was further applied to draw comparison with experimental measurements of net pressure gain for identical flow conditions. Only one of the two tested injector configurations followed the predicted trends, suggesting that performance of the second was governed by physical processes other than the reactant thermodynamics. Although an absolute pressure gain was not demonstrated, it is promising that the natural gas–air RDE delivered up to 90% of the theoretical performance.

20 FOSSIL-FUELED POWER PLANTS↗
NASA NTRS1993

Heat pipe cooled heat rejection subsystem modelling for nuclear electric propulsion

NASA LeRC is currently developing a FORTRAN based computer model of a complete nuclear electric propulsion (NEP) vehicle that can be used for piloted and cargo missions to the Moon or Mars. Proposed designs feature either a Brayton or a K-Rankine power conversion cycle to drive a turbine coupled with rotary alternators. Both ion and magnetoplasmodynamic (MPD) thrusters will be considered in the model. In support of the NEP model, Rocketdyne is developing power conversion, heat rejection, and power management and distribution (PMAD) subroutines. The subroutines will be incorporated into the NEP vehicle model which will be written by NASA LeRC. The purpose is to document the heat pipe cooled heat rejection subsystem model and its supporting subroutines. The heat pipe cooled heat rejection subsystem model is designed to provide estimate of the mass and performance of the equipment used to reject heat from Brayton and Rankine cycle power conversion systems. The subroutine models the ductwork and heat pipe cooled manifold for a gas cooled Brayton; the heat sink heat exchanger, liquid loop piping, expansion compensator, pump and manifold for a liquid loop cooled Brayton; and a shear flow condenser for a K-Rankine system. In each case, the final heat rejection is made by way of a heat pipe radiator. The radiator is sized to reject the amount of heat necessary.

Moriarty, Michael P.↗
NASA NTRS2018

Pulsed Fission-Fusion (PuFF)

In September 2013 the NASA Innovative Advanced Concept (NIAC) organization awarded a phase I contract to the PuFF team. Our phase 1 proposal discussed a pulsed fission-fusion propulsion system that injected gaseous deuterium (D) and tritium (T) as a mixture in a column, surrounded concentrically by gaseous uranium fluoride (UF6) and then an outer shell of liquid lithium. A high power current would flow down the liquid lithium and the resulting Lorentz force would compress the column by roughly a factor of 10. The compressed column would reach criticality and a combination of fission and fusion reactions would occur. The fission reactions would further energize the fusion center, and the fusion reactions would generate neutrons that promote more complete burnup of the fission fuel. The lithium liner provides some help as a neutron reflector but also acts as a propulsive medium, being converted to plasma which is then expanded against a magnetic nozzle for thrust. The expansion of the (primarily) lithium plasma against the nozzle's magnetic field inducts a current that is used to charge the system for the next pulse. Our concept also included secondary injection of a Field Reversed Configuration (FRC) plasmoid that would provide a secondary compression direction, axially against the column, and push the column away from the injection manifold, increasing the manifold's survivability.Our phase 1 proposal included modeling the above process first under steady state assumptions and second under a time variant integration. We proposed including these results into a Mars concept vehicle and finally proposing promising conditions to be evaluated experimentally in Phase II. In phase I we quickly realized that we needed to modify our approach. Our steady state work was completed as proposed, and the results indicated that one, a two stage compression system was not needed and two, that we wanted to move away from UF6. The steady state model shows much more margin than expected, to the point that we may well reach breakeven with the Charger – 1 facility, a 572 kJ Marx bank currently under refurbishment at UAH. Additionally we found that using gaseous D-T and UF6, provided a relatively simple prospect of using a pulsed injector, made reaching criticality more difficult. The introduction of large amounts of fluorine meant a radiative sink, sapping power from the fusion plasma and was harder to handle. Therefore we moved to a solid uranium target that held D-T under pressure. In so doing we could move our target closer to criticality and remove any material that did not sustain the reaction.

Fusion↗
DOE OSTI2020

An a priori evaluation of a principal component and artificial neural network based combustion model in diesel engine conditions

A principal component analysis (PCA) and artificial neural network (ANN) based chemistry tabulation approach is presented. ANNs are used to map the thermochemical state onto a low-dimensional manifold consisting of five control variables that have been identified using PCA. Three canonical configurations are considered to train the PCA-ANN model: a series of homogeneous reactors, a nonpremixed flamelet, and a two-dimensional lifted flame. The performance of the model in predicting the thermochemical manifold of a spatially-developing turbulent jet flame in diesel engine thermochemical conditions is a priori evaluated using direct numerical simulation (DNS) data. The PCA-ANN approach is compared with a conventional tabulation approach (tabulation using ad hoc defined control variables and linear interpolation). The PCA-ANN model provides higher accuracy and requires several orders of magnitude less memory. Here, these observations indicate that the PCA-ANN model is superior for chemistry tabulation, especially for modelling complex chemistries that present multiple combustion modes as observed in diesel combustion. The performance of the PCA-ANN model is then compared to the optimal estimator, i.e. the conditional mean from the DNS. The results indicate that the PCA-ANN model gives high prediction accuracy, comparable to the optimal estimator, especially for major species and the thermophysical properties. Higher errors are observed for the minor species and reaction rate predictions when compared to the optimal estimator. It is shown that the prediction of minor species and reaction rates can be improved by using training data that exhibits a variation of parameters as observed in the turbulent flame. The output of the ANN is analysed to assess mass conservation. It is observed that the ANN incurs a mean absolute error of 0.05% in mass conservation. Furthermore, it is demonstrated that this error can be reduced by modifying the cost function of the ANN to penalise for deviation from mass conservation.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Geometric prior of multi-resolution yielding manifolds and the local closest point projection for nearly non-smooth plasticity

Elastoplasticity models often introduce a scalar-valued yield function to implicitly represent the boundary between elastic and plastic material states. Here this paper introduces a new alternative where the yield envelope is represented by a manifold of which the topology and the geometry are learned from a set of data points in a parametric space (e.g. principal stress space, π-plane). Here, deep geometric learning enables us to reconstruct a highly complex yield envelope by breaking it down into multiple coordinate charts. The global atlas that consists of these coordinate charts in return allows us to represent the yield surface via multiple overlapping patches, each with a specific local parametrization. This setup provides several advantages over the classical implicit function representation approach. For instance, the availability of coordinate charts enables us to introduce an alternative stress integration algorithm where the trial stress may project directly on a local patch and hence circumvent the issues related to non-smoothness and the lack of convexity of yield surfaces. Meanwhile, the local parametric approach also enables us to predict hardening/softening locally in the parametric space, even without complete knowledge of the yield surface. Comparisons between the classical yield function approach on the non-smooth plasticity and anisotropic cam-clay plasticity model are provided to demonstrate the capacity of the models for highly precise yield surface and the feasibility of the implementation of the learned model in the local stress integration algorithm.

42 ENGINEERING↗
DOE OSTI2022

AI and extreme scale computing to learn and infer the physics of higher order gravitational wave modes of quasi-circular, spinning, non-precessing black hole mergers

We use artificial intelligence (AI) to learn and infer the physics of higher order gravitational wave modes of quasi-circular, spinning, non precessing binary black hole mergers. We trained AI models using 14 million waveforms, produced with the surrogate model NRHybSur3dq8, that include modes up to $\ell$ ≤ 4 and (5,5), except for (4,0) and (4,1), that describe binaries with mass-ratios $\textit{q}$ ≤ 8, individual spins $s^z_{\{1,2\}} \in$[–0.8,0.8], and inclination angle $θ \in$ [0,π]. Our probabilistic AI surrogates can accurately constrain the mass-ratio, individual spins, effective spin, and inclination angle of numerical relativity waveforms that describe such signal manifold. We compared the predictions of our AI models with Gaussian process regression, random forest, k-nearest neighbors, and linear regression, and with traditional Bayesian inference methods through the PyCBC Inference toolkit, finding that AI outperforms all these approaches in terms of accuracy, and are between three to four orders of magnitude faster than traditional Bayesian inference methods. Our AI surrogates were trained within 3.4 hours using distributed training on 1,536 NVIDIA V100 GPUs in the Summit supercomputer.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

Short-Wave Infrared Upconverting Nanoparticles

Optical technologies enable real-time, noninvasive analysis of complex systems but are limited to discrete regions of the optical spectrum. While wavelengths in the short-wave infrared (SWIR) window (typically, 1700-3000 nm) should enable deep subsurface penetration and reduced photodamage, there are few luminescent probes that can be excited in this region. Here, we report the discovery of lanthanide-based upconverting nanoparticles (UCNPs) that efficiently convert 1740 or 1950 nm excitation to wavelengths compatible with conventional silicon detectors. Screening of Ln3+ ion combinations by differential rate equation modeling identifies Ho3+/Tm3+ or Tm3+ dopants with strong visible or NIR-I emission following SWIR excitation. Experimental upconverted photoluminescence excitation (U-PLE) spectra find that 10% Tm3+-doped NaYF4 core/shell UCNPs have the strongest 800 nm emission from SWIR wavelengths, while UCNPs with an added 2% or 10% Ho3+ show the strongest red emission when excited at 1740 or 1950 nm. Mechanistic modeling shows that addition of a low percentage of Ho3+ to Tm3+-doped UCNPs shifts their emission from 800 to 652 nm by acting as a hub of efficient SWIR energy acceptance and redistribution up to visible emission manifolds. Parallel experimental and computational analysis shows rate equation models are able to predict compositions for specific wavelengths of both excitation and emission. These SWIR-responsive probes open a new IR bioimaging window, and are responsive at wavelengths important for vision technologies.

Qi, Xiao↗
DOE OSTI2023

Manipulating Growth and Propagation of Correlations in Dipolar Multilayers: From Pair Production to Bosonic Kitaev Models

Here, we study the nonequilibrium dynamics of dipoles confined in multiple stacked two-dimensional layers realizing a long-range interacting quantum spin 1/2 XXX model. We demonstrate that strong in-plane interactions can protect a manifold of collective layer dynamics. This then allows us to map the many-body spin dynamics to bosonic models. In a bilayer configuration we show how to engineer the paradigmatic two-mode squeezing Hamiltonian known from quantum optics, resulting in exponential production of entangled pairs and generation of metrologically useful entanglement from initially prepared product states. In multilayer configurations we engineer a bosonic variant of the Kitaev model displaying chiral propagation along the layer direction. Our study illustrates how the control over interactions, lattice geometry, and state preparation in interacting dipolar systems uniquely afforded by AMO platforms such as Rydberg and magnetic atoms, polar molecules, or trapped ions allows for the control over the temporal and spatial propagation of correlations for applications in quantum sensing and quantum simulation.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2022

Tri-Sectional Approximation of the Shortest Path to Long-Term Voltage Stability Boundary with Distributed Energy Resources

Ensuring long-term voltage stability is critical for reliable operations of power grids. High share of distributed energy resources (DERs) can create complicated system operation modes that may invalidate the traditional long-term voltage stability analysis based on typical operation modes. To address this challenge, this paper investigates how to compute the shortest path to the voltage stability boundary in the DER aggregated load space with large dispersion. Instead of working in the Euclidean space, we establish the analysis and computations on the algebraic power flow manifold to better capture the curvature change of the shortest path along the direction of losing stability. A modified optimal control framework is presented for obtaining the ground-truth of the smooth shortest path on the manifold. To efficiently and accurately solve for the shortest path, we further leverage the geometric features of the power flow manifold and propose a tri-sectional approximation model that is scalable for large-scale systems. Several numerical examples, up to the 1354-bus system, with different DER penetration levels and high dimensional renewable power injection variations are evaluated. The simulation results demonstrate that the tri-sectional approximation achieves high accuracy and efficiency to approximate the shortest path to the voltage stability boundary.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2024

tLaSDI: Thermodynamics-informed latent space dynamics identification

Here we propose a latent space dynamics identification method, namely tLaSDI, that embeds the first and second principles of thermodynamics. The latent variables are learned through an autoencoder as a nonlinear dimension reduction model. The latent dynamics are constructed by a neural network-based model that precisely preserves certain structures for the thermodynamic laws through the GENERIC formalism. An abstract error estimate is established, which provides a new loss formulation involving the Jacobian computation of autoencoder. The autoencoder and the latent dynamics are simultaneously trained to minimize the new loss. Computational examples demonstrate the effectiveness of tLaSDI, which exhibits robust generalization ability, even in extrapolation. In addition, an intriguing correlation is empirically observed between a quantity from tLaSDI in the latent space and the behaviors of the full-state solution.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Solvent effects on triplet yields in BODIPY-based photosensitizers

Here, we employ molecular dynamics simulations and quantum rate theories to elucidate the complex condensed-phase dynamics underpinning triplet-state formation in organic photosensitizers. Using models informed by first-principles calculations complete with a molecular representation of solvents of different polarities, we elucidate the interplay of the internal and environmental interactions underlying triplet yield. We find that triplet yields depend sensitively on the dielectric stabilization of the charge transfer intermediate that facilitates a transition into the triplet manifold. Our results illustrate the importance of molecularly detailed models in understanding the excited-state internal charge-transfer dynamics of photochemically relevant organic molecules.

Coello Escalante, Leonardo [University of Californ↗
DOE OSTI2024

Diabatic Valence-Hole Concept

A global diabatization scheme, based on the “valence-hole” concept, has been previously applied to model webs of avoided crossings that exist in four electronic-state symmetry manifolds of C 2 ( 1 Π g , 3 Π g , 1 Σ u + , and 3 Σ u + ). Here, this model is extended to the electronically excited states of four more molecules: CN ( 2 Σ + ), N 2 ( 3 Π u ), SiC ( 3 Π), and Si 2 ( 3 Π g ). Many strangenesses in the spectroscopic observations (e.g., energy level structure, predissociation linewidths, and radiative lifetimes) for all four electronic state systems discussed here are accounted for by this unified model. The key concept of the model is valence-hole electron configurations: 3σ 2 4σ 1 1π 4 5σ 2 in CN ( 2 Σ + ), 2σ g 2 2σ u 1 1π u 4 3σ g 2 1π g 1 in N 2 ( 3 Π u ), 5σ 2 6σ 1 7σ 2 2π 3 in SiC ( 3 Π), and 4σ g 2 4σ u 1 5σ g 2 2π u 3 in Si 2 ( 3 Π g ), all of which have a triply occupied “valence-core” (i.e., 2σ g 2 2σ u 1 or the equivalent). These valence-hole configurations have a nominal bond order of three or higher and correlate with high-energy separated-atom limits with an np ← ns (n = 2, 3) promotion in one of the atomic constituents. On its way to dissociation, the strongly bound diabatic valence-hole state crosses multiple weakly bound or repulsive states, which belong to electron configurations with a completely filled valence-core. These curve crossings between diabatic potentials result in a network of many avoided crossings among multiple electronic states, analogous to the well-studied electronic structure landscape of ionic-covalent crossings in strongly ionic molecules. In conclusion, considering the unique role of valence-hole states in shaping the global electronic structure, the valence-hole concept should be added to our intuitive framework of chemical bonding.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗