Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “JuMP 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 127 records · Page 7
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Multiclass Classification Using Bayesian Multivariate Adaptive Regression Splines

We present a new Bayesian model for the problem of multiclass classification. In this model, the probabilities of class membership of a given observation are determined by the mean of a latent Gaussian distribution. The mean functions of this latent distribution consist of combinations of highly flexible basis functions of the inputs: multivariate adaptive regression splines (MARS), first developed for multiple regression. We use reversible jump Markov chain Monte Carlo to make inference on the classification model, including the number of basis functions. We compare the probabilistic classification performance of our proposed approach to existing methods on simulated and benchmark data, and compare uncertainty estimates on simulated data. Our proposed method compares favorably with existing Bayesian and frequentist multiclass classification methods in out-of-sample probabilistic classification, and uncertainty estimation of these probabilistic classifications. We examine the fit of the proposed method to a data set of hurricane storm surge levels near Delaware Bay, US, and conclude that sea level rise is a key contributor to damage delivered by storm surge.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Adaptive Interface-PINNs (AdaI-PINNs): An Efficient Physics-Informed Neural Networks Framework for Interface Problems

Here, we present an efficient physics-informed neural networks (PINNs) framework, termed Adaptive Interface-PINNs (AdaI-PINNs), to improve the modeling of interface problems with discontinuous coefficients and/or interfacial jumps. This framework is an enhanced version of its predecessor, Interface PINNs or I-PINNs (Sarma et al.; https://doi.org/10.1016/j.cma.2024.117135), which involves domain decomposition and assignment of different predefined activation functions to the neural networks in each subdomain across a sharp interface, while keeping all other parameters of the neural networks identical. In AdaI-PINNs, the activation functions vary solely in their slopes, which are trained along with the other parameters of the neural networks. This makes the AdaI-PINNs framework fully automated without requiring preset activation functions. Comparative studies on one-dimensional, two-dimensional, and three-dimensional benchmark elliptic interface problems reveal that AdaI-PINNs outperform I-PINNs, reducing computational costs by 2-6 times while producing similar or better accuracy.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Summary of GFL Inverter Capability and Performance: Malawi [Slides]

This presentation provides a comprehensive comparison of inverter-based resource (IBR) technical requirements across multiple international grid codes - including IEEE 2800, ESO, EirGrid, and Malawi's evolving regulatory framework - to support capacity building for electricity-sector institutions in Malawi. It highlights key performance expectations for grid-following (GFL) inverters across critical domains such as primary and fast frequency response, ramping behavior, frequency and voltage ride-through capabilities, reactive power and voltage control, ROCOF withstand, phase jump tolerance, system strength considerations, protection schemes, and dynamic modeling requirements. Emphasizing a "Do no harm, do something, do more" philosophy, the document synthesizes best practices and operational benchmarks to guide the integration and stability of inverter-based resources within Malawi's power system. The presentation forms part of the SABESS Center of Excellence initiative supported by the Global Energy Alliance for People and the Planet.

24 POWER TRANSMISSION AND DISTRIBUTION↗
NASA NTRS1985

Colloidal Oxide Properties and Kinetic Studies

Silica based oxide particles prepared by various methods are compared in terms of their gellation kinetics, densification kinetics, and microable structural evolutions begin at very low temperature. Additional work reported in literature further suggested that the colloidal behaviors and the dried gel structures of the same oxide from different preparation techniques differ fundamentally. These differences significantly affect the densification kinetics and the product microstructure. Gravitational segregation of particles of different sizes in a gravitational field is modelled as a diffusion/segregation problem. A structural free volume model was used to describe the activation process of jump statistics. The sizes of the particles, the interstices, and their packing topology are accounted for in the estimation of the free volume. An attempt frequency is ascribed to particles to account for the vibrational rearrangement during powder packing.

Chen, I. W.↗
NASA NTRS2018

Lessons for WFIRST CGI from Ground-Based High-Contrast Systems

The Coronagraph Instrument (CGI) for NASA’s Wide Field Infrared Survey Telescope (WFIRST) will constitute a dramatic step forward for high-contrast imaging, integral field spectroscopy, and polarimetry of exoplanets and circumstellar disks, aiming to improve upon the sensitivity of current ground-based direct imaging facilities by 2–3 orders of magnitude. Furthermore, CGI will serve as a pathfinder for future exo-Earth imaging and characterization missions by demonstrating wavefront control, coronagraphy, and spectral retrieval in a new contrast regime, and by validating instrument and telescope models at unprecedented levels of precision. To achieve this jump in performance, it is critical to draw on the experience of ground-based high-contrast facilities. We discuss several areas of relevant commonalities, including: wavefront control, post-processing of integral field unit data, and calibration and observing strategies.

Xuan, Wenhao↗
NASA NTRS2015

Application of the Refined Zigzag Theory to the Modeling of Delaminations in Laminated Composites

The Refined Zigzag Theory is applied to the modeling of delaminations in laminated composites. The commonly used cohesive zone approach is adapted for use within a continuum mechanics model, and then used to predict the onset and propagation of delamination in five cross-ply composite beams. The resin-rich area between individual composite plies is modeled explicitly using thin, discrete layers with isotropic material properties. A damage model is applied to these resin-rich layers to enable tracking of delamination propagation. The displacement jump across the damaged interfacial resin layer is captured using the zigzag function of the Refined Zigzag Theory. The overall model predicts the initiation of delamination to within 8% compared to experimental results and the load drop after propagation is represented accurately.

Groh, Rainer M. J.↗
NASA NTRS1988

Effects of compressibility on the temperature jump at the interface of layered, spherical-shell convection

Large temperature jumps at the interface of layered convection are important to the argument used against the likelihood of separate circulations in the upper and lower mantles. This problem was studied within the framework of a compressible, constant viscosity spherical-shell model. Both mechanical and thermal coupling configurations are considered. Although the temperature jumps are reduced by compressibility, their magnitudes remain quite large, in the case of mechanical coupling. For thermal coupling, the temperature jumps become smaller but still are substantial, between 500 to 1000 C. In layered spherical-shell convection, flows in the lower mantle are several times greater than the surface velocities.

Yen, David A.↗
NASA NTRS1998

1-D Transient Thermal Modeling of an Ablative Material (MCC-1) Exposed to a Simulated Convective Titan 4 Launch Environment

The purpose of the work is to demonstrate that the flat test panel substrate temperatures are consistent with analysis predictions for MCC-1 applied to a aluminum substrate. The testing was performed in an aerothermal facility on samples of three different thicknesses of MCC-1 on an aluminum substrate. The results of the test were compared with a Transient Thermal model. The key assumptions of the Transient Thermal model were: (1) a one-dimensional heat transfer; (2) a constant ablation recession rate (determined from pre and post-test measurements); (3) ablation temperature of 540 degrees F; (4) Char left behind the ablation front; and (5) temperature jump correction for incident heat transfer coefficient. Two methods were used to model the heating of bare MCC-1: (1) Directly input surface temperature as a function of time; and (2) Aerothermal heating using calibration plate data and subtracting the radiation losses to tunnel walls. The results are presented as graphs. This article is presented in Viewgraph format.

Reinarts, Thomas R.↗
NASA NTRS1979

Thermal tides and Martian dust storms - Direct evidence for coupling

Evidence for the coupling of diurnal and semidiurnal thermal tides, observed as surface pressure oscillations at the Viking Lander 1 and 2 landing sites, to Martian global dust storms is presented and the implications for models of dust storm generation are discussed. Local atmospheric opacity increases and jumps in atmospheric pressure ranges were found to be essentially simultaneous, and diurnal and semidiurnal components of the pressure and wind oscillations were observed to increase markedly after the onset of a dust storm. The semidiurnal wind component can be related to the observed surface pressure variations by means of a model of a semidiurnally forced Ekman boundary layer. However, a classical atmospheric tidal model shows that the preferential enhancement of the semidiurnal surface pressure oscillation observed at the Lander 1 site can be produced by a tidal heating distribution in which most of the heating is done at an altitude above 10 km. Observations suggest that various mechanisms are important for the generation and decay of global Martian dust storms.

Leovy, C. B.↗
DOE OSTI2023

CIEL ∗ Ch color map for visualization and analysis of sea ice motion

The International Commission on Illumination (CIE) designed its color space to be perceptually uniform so that a given numerical change in the color code corresponds to perceived change in color. This color encoding is demon- strated to be advantageous in scientific visualization and analysis of vector fields. The specific application is analysis of ice motion in the Arctic where patterns in smooth monthly-averaged ice motion are seen. Furthermore, fractures occurring in the ice cover result in discontinuities in the ice motion. This vector jump in displacement can also be visualized. We then analyze modeled and observed fractures through the use of a metric on the color space, and image amplitude and phase metrics. Amplitude and phase met- rics arise from image registration that is accomplished by sampling images using space filling curves, thus reducing the image registration problem to the more reliable functional alignment problem. We demonstrate this through an exploration of the metrics to compare model runs to an observed ice crack.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1987

Revised NASA axially symmetric ring model for coupled-cavity traveling-wave tubes

A versatile large-signal, two-dimensional computer program is used by NASA to model coupled-cavity travelling-wave tubes (TWTs). In this model, the electron beam is divided into a series of disks, each of which is further divided into axially symmetric rings which can expand and contract. The trajectories of the electron rings and the radiofrequency (RF) fields are determined from the calculated axial and radial space-charge, RF, and magnetic forces as the rings pass through a sequence of cavities. By varying electrical and geometric properties of individual cavities, the model is capable of simulating severs, velocity tapers, and voltage jumps. The calculated electron ring trajectories can be used in designing magnetic focusing and multidepressed collectors. The details of using the program are presented, and results are compared with experimental data.

Wilson, Jeffrey D.↗
NASA NTRS1987

Computational fluid dynamics and aerothermodynamics

At the present time the efforts on this project are organized into two areas: the development of an approximate stagnation point solution and approximate flowfield studies which can be used to develop and investigate shock jump, electron temperature, radiation, vibration-dissociation coupling, and chemistry models. Progress in each area is discussed.

Carlson, Leland A.↗
NASA NTRS1991

Shock wave interaction with an abrupt area change

The wave patterns that occur when a shock wave interacts with an abrupt area changed are analyzed in terms of the incident shock wave Mach number and area-jump ratio. The solutions predicted by a semi-similar models are in good agreement with those obtained numerically from the quasi-one-dimensional time-dependent Euler equations. The entropy production for the wave system is defined and the principle of minimum entropy production is used to resolve a nonuniqueness problem of the self-similar model.

Salas, Manuel D.↗
NASA NTRS1990

Simulation of period doubling of recurrent solar wind structures

Based on satellite observations of a recurrent solar wind structure conducted in 1974, an MHD simulation model, and input functions generated from plasma and magnetic field data, the continuing evolution of the solar wind structure outside 5 AU is studied. The model uses the Rankine-Hugoniot relations to describe the jumps in flow properties across the shocks, and it treats shocks as surfaces of discontinuity with zero thickness. Two interaction processes (the collision and the merging of shocks) play important roles in restructuring the solar wind in the outer heliosphere. The simulation result shows that period doubling occurs between 5 and 10 AU. The recurrent solar wind appears to be a persistent new structure between 10 and 20 AU, and it consists of one merged interaction region per solar rotation.

Whang, Y. C.↗
DOE OSTI2023

CoolPINNs: A physics-informed neural network modeling of active cooling in vascular systems

Emerging technologies like hypersonic aircraft, space exploration vehicles, and batteries avail fluid circulation in embedded microvasculatures for efficient thermal regulation. Modeling is vital during the design and operational phases of these engineered systems. However, many challenges exist in developing a modeling framework. What is lacking is an accurate framework that (i) captures sharp jumps in the thermal flux across complex vasculature layouts, (ii) deals with oblique derivatives (involving tangential and normal components), (iii) handles nonlinearity because of radiative heat transfer, (iv) provides a high-speed forecast for real-time monitoring, and (v) facilitates robust inverse modeling. Here, this paper addresses these challenges by availing the power of physics-informed neural networks (PINNs). We develop a fast, reliable, and accurate Scientific Machine Learning (SciML) framework for vascular-based thermal regulation—called CoolPINNs: a PINNs-based modeling framework for active cooling. The proposed mesh-less framework elegantly overcomes all the mentioned challenges. The significance of the reported research is multi-fold. First, the framework is valuable for real-time monitoring of thermal regulatory systems because of rapid forecasting. Second, researchers can address complex thermoregulation designs since the approach is meshless. Finally, the framework facilitates systematic parameter identification and inverse modeling studies, perhaps the most significant utility of the current framework.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

ZND Verification Tests for Reactive Burn Models in FLAG

The ZND theory, named after Zeldovich, von Neumann, and Doering, provides a simple model for one-dimensional ideal steady-state detonation. It assumes that the detonation wave front starts with a shock that is a discontinuous jump and is followed by a finite-length reaction zone. Reactive burn (also called reactive flow) models are based on ZND theory, as they model the shock initiation and detonation process with a finite reaction rate. The ZND wave propagation test is essentially the only available test case where an analytic solution exists for verification of reactive burn models in numerical codes. However, there are extensions and variants of the ZND test that have been devised for verification of multidimensional flows. The objective of this work is to provide verification of the reactive burn models currently implemented in the Lagrangian hydrocode FLAG and investigate the influence of mesh resolution, artificial viscosity models, and the Arbitrary Lagrangian-Eulerian (ALE) Euler relaxer on the simulation results. The burn models of interest are the Wescott-Stewart-Davis (WSD) model, the Scaled Uniform Reactive Flow (SURF) model (specifically with the SURFplus model extension), and the Arrhenius shock temperature state dependent WSD (AWSD) model. Previously, Ralph Menikoff has used ZND tests for verification of the SURF and SURFplus models in the Eulerian hydrocode xRAGE. The ZND tests here are somewhat different than the approach by Menikoff. In particular, we use a piston-driven ZND detonation wave (via a prescribed constant velocity boundary condition) in a Lagrangian framework whereas Menikoff had a ZND wave followed by a invariant rarefaction wave. The xRAGE simulations were carried out on uniform grids and adaptive mesh refinement (AMR) grids. Although AMR was recently implemented in FLAG and now fully functional for 2D simulations, it will not be evaluated with ZND tests at this time. Some work has been done previously for validating the reactive burn models in FLAG. For example, the validation studies of SURF with shock-to-detonation (SDT) tests, cylinder tests, and gap-stick tests. Further validation of the AWSD, WSD, and SURF models is described in for SDT, multi-shock, cylinder, and corner-turning tests. Recently, a large parameter study with approximately twenty SDT tests was performed to validate the AWSD, WSD, SURF, and SURFplus models while also investigating mesh resolution and artificial viscosity settings. To the best of our knowledge, the current work represents the first documented verification of these burn models in FLAG. The reader should note that many of the tables and figures in this report use units of cm/g/µs, which are the default units for FLAG. However, some lengths and velocities are expressed in µm and mm/µs, respectively, which are typical units for detonation analyses.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗
NASA NTRS2009

Simulation of RCC Crack Growth Due to Carbon Oxidation in High-Temperature Gas Environments

The carbon wall oxidation technique coupled with a CFD technique was employed to study the flow in the expanding crack channel caused by the oxidation of the channel carbon walls. The recessing 3D surface morphing procedure was developed and tested in comparison with the arcjet experimental results. The multi-block structured adaptive meshing was used to model the computational domain changes due to the wall recession. Wall regression rates for a reinforced carbon-carbon (RCC) samples, that were tested in a high enthalpy arcjet environment, were computationally obtained and used to assess the channel expansion. The test geometry and flow conditions render the flow regime as the transitional to continuum, therefore Navier-Stokes gas dynamic approach with the temperature jump and velocity slip correction to the boundary conditions was used. The modeled mechanism for wall material loss was atomic oxygen reaction with bare carbon. The predicted channel growth was found to agree with arcjet observations. Local gas flow field results were found to affect the oxidation rate in a manner that cannot be predicted by previous mass loss correlations. The method holds promise for future modeling of materials gas-dynamic interactions for hypersonic flight.

Titov, E. V.↗
NASA NTRS2005

The 0.125 degree finite-volume General Circulation Model on the NASA Columbia Supercomputer: Preliminary Simulations of Mesoscale Vortices

The NASA Columbia supercomputer was ranked second on the TOP500 List in November, 2004. Such a quantum jump in computing power provides unprecedented opportunities to conduct ultra-high resolution simulations with the finite-volume General Circulation Model (fvGCM). During 2004, the model was run in realtime experimentally at 0.25 degree resolution producing remarkable hurricane forecasts [Atlas et al., 2005]. In 2005, the horizontal resolution was further doubled, which makes the fvGCM comparable to the first mesoscale resolving General Circulation Model at the Earth Simulator Center [Ohfuchi et al., 2004]. Nine 5-day 0.125 degree simulations of three hurricanes in 2004 are presented first for model validation. Then it is shown how the model can simulate the formation of the Catalina eddies and Hawaiian lee vortices, which are generated by the interaction of the synoptic-scale flow with surface forcing, and have never been reproduced in a GCM before.)

Shen, B.-W.↗