SEARCH · Engineering Papers
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.
The method circumvents many of the existing difficulties in computational logic presently encountered in the direct analytical or numerical evaluation of the appropriate surface integral. It may be applied to complex spacecraft structures for computing the total force arising from either specular or diffuse reflection or even from non-Lambertian reflection and re-radiation.
The current practice for fusion scientists running first principle simulations on high performance computing plat-forms is to either run their simulations and output their data for post-hoc analysis, or to place in situ analytics into their code. In this paper we examine a complex workflow using XGC fusions simulation run on the Oak Ridge Leadership Computing Facility's supercomputer Summit, which also involve three anal-yses as part of the results necessary for scientific discovery. We discuss the challenges faced when implementing these algorithms and present an original hybrid staging technique to help enable the physicists to make discoveries during the execution of the simulation. By creating this infrastructure, we can examine complicated physics results, which may not have been possible without the infrastructure. For example, our work enables the online visualization of turbulent homoclinic tangle around the magnetic X-point, breaking the last confinement surface. This visualization could help fusion scientists to better understand and improve the turbulence spread of plasma exhaust heat, which is crucial toward realizing plasmas beyond the currently accessible physics regimes of present-day tokamak reactors. The physics of turbulent homoclinic tangle will be reported in a future physics publication, by utilizing the original online analysis/visualization framework presented in this paper.
Not Available
The thrust of the proposal was to exploit modern data-mining tools in a way that will create a systematic, computer-assisted approach to the representation of random media -- and also to the representation of the solutions of an array of important physicochemical processes that take place in/on such media. A parsimonious representation/parametrization of the random media links directly (via uncertainty quantification tools) to good sampling of the distribution of random media realizations. It also links directly to modern multiscale computational algorithms (like the equation-free approach that has been developed in our group) and plays a crucial role in accelerating the scientific computation of solutions of nonlinear PDE models (deterministic or stochastic) in such media – both solutions in particular realizations of the random media, and estimation of the statistics of the solutions over multiple realizations (e.g. expectations).
The accurate tracking of vortex evolution from Large Eddy Simulation (LES) data is a complex and computationally intensive problem. The vortex tracking requires the analysis of very large three-dimensional and time-varying datasets. The complexity of the problem is further compounded by the fact that these vortices are embedded in a background turbulence field, and they may interact with the ground surface. Another level of complication can arise, if vortices from multiple aircrafts are simulated. This paper presents a new technique for post-processing LES data to obtain wake vortex tracks and wake intensities. The new approach isolates vortices by defining "regions of interest" (ROI) around each vortex and has the ability to identify vortex pairs from multiple aircraft. The paper describes the new methodology for tracking wake vortices and presents application of the technique for single and multiple aircraft.
The flow field over an airfoil contains several distinct flow characteristics, e.g. laminar, transitional, turbulent boundary layer flow, flow separation, unstable free shear layers, and a wake. This diversity of flow regimes taxes the presently available Reynolds averaged turbulence models. Such models are generally tuned to predict a particular flow regime, and adjustments are necessary for the prediction of a different flow regime. Similar difficulties are likely to emerge when the large eddy simulation technique is applied with the widely used Smagorinsky model. This model has not been successful in correctly representing different turbulent flow fields with a single universal constant and has an incorrect near-wall behavior. Germano et al. (1991) and Ghosal, Lund & Moin have developed a new subgrid-scale model, the dynamic model, which is very promising in alleviating many of the persistent inadequacies of the Smagorinsky model: the model coefficient is computed dynamically as the calculation progresses rather than input a priori. The model has been remarkably successful in prediction of several turbulent and transitional flows. We plan to simulate turbulent flow over a '2D' airfoil using the large eddy simulation technique. Our primary objective is to assess the performance of the newly developed dynamic subgrid-scale model for computation of complex flows about aircraft components and to compare the results with those obtained using the Reynolds average approach and experiments. The present computation represents the first application of large eddy simulation to a flow of aeronautical interest and a key demonstration of the capabilities of the large eddy simulation technique.
A new method of domain decomposition has been developed for generating unstructured grids in subdomains either sequentially or using multiple computers in parallel. Domain decomposition is a crucial and challenging step for parallel grid generation. Prior methods are generally based on auxiliary, complex, and computationally intensive operations for defining partition interfaces and usually produce grids of lower quality than those generated in single domains. The new technique, referred to as "Advancing Partition," is based on the Advancing-Front method, which partitions a domain as part of the volume mesh generation in a consistent and "natural" way. The benefits of this approach are: 1) the process of domain decomposition is highly automated, 2) partitioning of domain does not compromise the quality of the generated grids, and 3) the computational overhead for domain decomposition is minimal. The new method has been implemented in NASA's unstructured grid generation code VGRID.
A new method of domain decomposition has been developed for generating unstructured grids in subdomains either sequentially or using multiple computers in parallel. Domain decomposition is a crucial and challenging step for parallel grid generation. Prior methods are generally based on auxiliary, complex, and computationally intensive operations for defining partition interfaces and usually produce grids of lower quality than those generated in single domains. The new technique, referred to as "Advancing Partition," is based on the Advancing-Front method, which partitions a domain as part of the volume mesh generation in a consistent and "natural" way. The benefits of this approach are: 1) the process of domain decomposition is highly automated, 2) partitioning of domain does not compromise the quality of the generated grids, and 3) the computational overhead for domain decomposition is minimal. The new method has been implemented in NASA's unstructured grid generation code VGRID.
The most relevant contractual tower design requirements and goal for the Mod-1 tower are related to steel truss tower construction, cost-effective state-of-the-art design, a design life of 30 years, and maximum wind conditions of 120 mph at 30 feet elevation. The Mod-1 tower design approach was an iterative process. Static design loads were calculated and member sizes and overall geometry chosen with the use of finite element computer techniques. Initial tower dynamic characteristics were then combined with the dynamic properties of the other wind turbine components, and a series of complex dynamic computer programs were run to establish a dynamic load set and then a second tower design.
Neutrinos in compact-object environments, such as core-collapse supernovae, can experience various kinds of collective effects in flavor space, engendered by neutrino-neutrino interactions. These include “bipolar” collective oscillations, which are exhibited by neutrino ensembles where different flavors dominate at different energies. Considering the importance of neutrinos in the dynamics and nucleosynthesis in these environments, it is desirable to ascertain whether an Earth-based detection could contain signatures of bipolar oscillations that occurred within a supernova envelope. To that end, we, in this study, continue examining a cost-function formulation of statistical data assimilation (SDA) to infer solutions to a small-scale model of neutrino flavor transformation. SDA is an inference paradigm designed to optimize a model with sparse data. Our model consists of two monoenergetic neutrino beams with different energies emanating from a source and coherently interacting with each other and with a matter background, with radially varying interaction strengths. We attempt to infer flavor transformation histories of these beams using simulated measurements of the flavor content at locations “in vacuum” (that is, far from the source), which could in principle correspond to Earth-based detectors. Within the scope of this small-scale model, we found that: (i) based on such measurements, the SDA procedure is able to infer whether bipolar oscillations had occurred within the protoneutron star envelope, and (ii) if the measurements sample the full amplitude of the neutrino oscillations in vacuum, then the amplitude of the prior bipolar oscillations is well predicted. This result intimates that the inference paradigm can well complement numerical integration codes, via its ability to infer flavor evolution at physically inaccessible locations.
A computer modeling tool is being developed to assess candidate designs for the Space Station Data Management System (DMS). The DMS is to be a complex distributed computer system including the processor, storage devices, local area networks, and software that will support all processing functions onboard the Space Station. The modeling tool will allow a candidate design for the DMS, or for other subsystems that use the DMS, to be evaluated in terms of parameters. The tool and its associated modeling methodology are intended for use by DMS and subsystem designers to perform tradeoff analyses between design concepts using varied architectures and technologies.
Context. With the recent launching of the Hitomi X-ray space observatory, K lines and edges of chemical elements with low cosmic abundances, namely F, Na, P, Cl, K, Sc, Ti, V, Cr, Mn, Co, Cu and Zn, can be resolved and used to determine important properties of supernova remnants, galaxy clusters and accreting black holes and neutron stars.Aims. The second stage of the present ongoing project involves the computation of the accurate photoabsorption and photoionisation cross sections required to interpret the X-ray spectra of such trace elements.Methods. Depending on target complexity and computer tractability, ground-state cross sections are computed either with the close-coupling Breit-Pauli R-matrix method or with the autostructure atomic structure code in the isolated-resonance approximation. The intermediate-coupling scheme is used whenever possible. In order to determine a realistic K-edge behaviour for each species, both radiative and Auger dampings are taken into account, the latter being included in the R-matrix formalism by means of an optical potential.Results. Photoabsorption and total and partial photoionisation cross sections are reported for isoelectronic sequences with electron numbers 3< or = N< or = 11. The Na sequence (N=11) is used to estimate the contributions from configurations with a 2s hole (i.e. [2s]) and those containing 3d orbitals, which will be crucial when considering sequences with N 11.Conclusions. It is found that the [2s/u] configurations must be included in the target representations of species with N> 11 as they contribute significantly to the monotonic background of the cross section between the L and K edges. Configurations with 3d orbitals are important in rendering an accurate L edge, but they can be practically neglected in the K-edge region.
In this work, we present a method for unsupervised learning of equations of motion for objects in raw and optionally distorted unlabeled synthetic video (or, more generally, for discovering and modeling predictable features in time-series data). We first train an autoencoder that maps each video frame into a low-dimensional latent space where the laws of motion are as simple as possible, by minimizing a combination of nonlinearity, acceleration, and prediction error. Differential equations describing the motion are then discovered using Pareto-optimal symbolic regression. We find that our pre-regression (“pregression”) step is able to rediscover Cartesian coordinates of unlabeled moving objects even when the video is distorted by a generalized lens. Using intuition from multidimensional knot theory, we find that the pregression step is facilitated by first adding extra latent space dimensions to avoid topological problems during training and then removing these extra dimensions via principal component analysis. An inertial frame is autodiscovered by minimizing the combined equation complexity for multiple experiments.
Not Available
The rapid growth of computer technology and innovation has played a significant role in the rise of computer automation of human tasks in modem production systems across all industries. Although the rationale for automation has been to eliminate "human error" or to relieve humans from manual repetitive tasks, various computer-related hazards and accidents have emerged as a direct result of increased system complexity attributed to computer automation. The risk assessment techniques utilized for electromechanical systems are not suitable for today's software-intensive systems or complex human-computer controlled systems.This thesis will propose a new systemic model-based framework for analyzing risk in safety-critical systems where both computers and humans are controlling safety-critical functions. A new systems accident model will be developed based upon modem systems theory and human cognitive processes to better characterize system accidents, the role of human operators, and the influence of software in its direct control of significant system functions Better risk assessments will then be achievable through the application of this new framework to complex human-computer controlled systems.
The rapid growth of computer technology and innovation has played a significant role in the rise of computer automation of human tasks in modem production systems across all industries. Although the rationale for automation has been to eliminate "human error" or to relieve humans from manual repetitive tasks, various computer-related hazards and accidents have emerged as a direct result of increased system complexity attributed to computer automation. The risk assessment techniques utilized for electromechanical systems are not suitable for today's software-intensive systems or complex human-computer controlled systems. This thesis will propose a new systemic model-based framework for analyzing risk in safety-critical systems where both computers and humans are controlling safety-critical functions. A new systems accident model will be developed based upon modem systems theory and human cognitive processes to better characterize system accidents, the role of human operators, and the influence of software in its direct control of significant system functions. Better risk assessments will then be achievable through the application of this new framework to complex human-computer controlled systems.
As we approach exascale, computational parallelism will have to drastically increase in order to meet throughput targets. Many–core architectures have exacerbated this problem by trading reduced clock speeds, core complexity, and computation throughput for increasing parallelism. This presents two major challenges for communication libraries such as MPI: the library must leverage the performance advantages of thread level parallelism and avoid the scalability problems associated with increasing the number of processes to that scale. Hybrid programming models, such as MPI+X, have been proposed to address these challenges. MPI THREAD MULTIPLE is MPI's thread safe mode. While there has been work to optimize it, it largely remains non–performant in most implementations. While current applications avoid MPI multithreading due to performance concerns, it is expected to be utilized in future applications. One of the major synchronous data structures required by MPI is the matching engine. In this paper, we present a parallel matching algorithm that can improve MPI matching for multithreaded applications. We then perform a feasibility study to demonstrate the performance benefit of the technique.
A generalization of the determinant appears in particle physics in effective Lagrangian interaction terms that model the chiral anomaly in quantum chromodynamics (Giacosa et al. in Phys Rev D 97(9):091901, 2018, Phys Rev D 109(7):L071502, 2024), in particular in connection to mesons. This polydeterminant function, known in the mathematical literature as a mixed discriminant, associates N distinct N x N complex matrices into a complex number and reduces to the usual determinant when all matrices are taken as equal. Here, we explore the main properties of the polydeterminant applied to (quantum) fields by using a formalism and a language close to high-energy physics approaches. We discuss its use as a tool to write down novel chiral anomalous Lagrangian terms and present an explicit illustrative model for mesons. Finally, the extension of the polydeterminant as a function of tensors is shown.