Search NASA⌕ Search

SEARCH · Search NASA

Results for “manifold sampling”

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 73 records · Page 4

Neural Active Manifolds: Nonlinear Dimensionality Reduction for Uncertainty Quantification

We present a new approach for nonlinear dimensionality reduction, specifically designed for computationally expensive mathematical models. We leverage autoencoders to discover a one-dimensional neural active manifold (NeurAM) capturing the model output variability, through the aid of a simultaneously learnt surrogate model with inputs on this manifold. Our method only relies on model evaluations and does not require the knowledge of gradients. The proposed dimensionality reduction framework can then be applied to assist outer loop many-query tasks in scientific computing, like sensitivity analysis and multifidelity uncertainty propagation. In particular, we prove, both theoretically under idealized conditions, and numerically in challenging test cases, how NeurAM can be used to obtain multifidelity sampling estimators with reduced variance by sampling the models on the discovered low-dimensional and shared manifold among models. Several numerical examples illustrate the main features of the proposed dimensionality reduction strategy and highlight its advantages with respect to existing approaches in the literature.

Autoencoders↗

Simulating sticky particles: A Monte Carlo method to sample a stratification

Many problems in materials science and biology involve particles interacting with strong, short-ranged bonds that can break and form on experimental timescales. Treating such bonds as constraints can significantly speed up sampling their equilibrium distribution, and there are several methods to sample probability distributions subject to fixed constraints. We introduce a Monte Carlo method to handle the case when constraints can break and form. More generally, the method samples a probability distribution on a stratification: a collection of manifolds of different dimensions, where the lower-dimensional manifolds lie on the boundaries of the higher-dimensional manifolds. We show several applications of the method in polymer physics, self-assembly of colloids, and volume calculation in high dimensions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Electronic and structural properties of RbCe X 2 ( X 2 : O 2 , S 2 , SeS, Se 2 , TeSe, Te 2 )

We report triangular lattice delafossite compounds built from magnetic lanthanide ions are a topic of recent interest due to their frustrated magnetism and realization of quantum disordered magnetic ground states. Here we report the evolution of the structure and electronic ground states of RbCeX 2 compounds, built from a triangular lattice of Ce 3+ ions, upon varying their anion character (X 2 =O 2 , S 2 , SeS, Se 2 , TeSe, Te 2 ). This includes the discovery of a new member of this series, RbCeO 2 , that potentially realizes a quantum disordered ground state analogous to NaYbO 2 . Magnetization and susceptibility measurements reveal that all compounds manifest mean-field antiferromagnetic interactions and, with the exception of the oxide, possess signatures of magnetic correlations onset below 1 K. The crystalline electric field level scheme is explored via neutron scattering and ab initio calculations in order to model the intramultiplet splitting of the J=5/2 multiplet. In addition to the two excited doublets expected within the J=5/2 manifold, we observe one extra local mode present across the sample series. This added mode shifts downward in energy with increasing anion mass and decreasing crystal field strength, suggesting a long-lived anomalous mode endemic to anion motion about the Ce3 + sites.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Measurements of dimethyl sulfide and H2S over the western North Atlantic and the tropical Atlantic

Airborne measurements of DMS and H2S were made off the east coast of the United States and over the tropical Atlantic off Brazil. Samples were collected through a fluorinated ethylene propylene Teflon inlet manifold. Dimethyl sulfide (DMS) was preconcentrated onto gold wool and analyzed by gas chromatography/flame photometric detection. H2S was collected on AgNO3-impregnated filters and determined by fluorescence quenching. Use of a new scrubber material (cotton) to remove negative interference on DMS measurements was investigated. Comparison with a Na2CO3/Anakrom scrubber gave good overall agreement. Only under extreme conditions, e.g., on flight 9 (continental air mass, low humidity, high O3, and low DMS values) did Na2CO3 show noticeable loss of DMS compared to cotton. On most flights, especially in marine air masses with high humidity and relatively low O3, the results from both scrubbers agreed well with each other and with other instruments used during the intercalibration. Off the U.S. East Coast, DMS levels showed strong dependence on air mass origin with high values (up to 83 ppt) in marine tropical air masses and low values (10-20 ppt) in continental and polar air. Over the tropical Atlantic, DMS ranged over 20-100 ppt in the mixed layer. Nighttime values were a factor of 1.6-2.3 higher than daytime levels. DMS decreased with altitude to less than 1 ppt at 4000 m. H2S in the mixed layer off the U.S. East Coast ranged from 10 to 200 ppt. Significant influence from terrestrial and pollution sources was evident. H2S in air masses originating over the eastern seaboard was much higher than in continental polar air or over the remote tropical continents. In contrast, over the tropical Atlantic, concentrations were very low (5-10 ppt), typical of truly marine air. Night/day ratios were about 1.4. No significant geographical variability was seen in H2S levels over the tropical Atlantic. The correlation of atmospheric Rn-222 and H2S was significant, with both being higher off the U.S. East Coast than over the tropical Atlantic.

Andreae, T. W.↗

Enhancing Model Predictability for a Scramjet Using Probabilistic Learning on Manifolds

In this study, the computational burden of a large-eddy simulation for reactive flows is exacerbated in the presence of uncertainty in flow conditions or kinetic variables. A comprehensive statistical analysis, with a sufficiently large number of samples, remains elusive. Statistical learning is an approach that allows for extracting more information using fewer samples. Such procedures, if successful, will greatly enhance the predictability of models in the sense of improving exploration and characterization of uncertainty due to model error and input dependencies, all while being constrained by the size of the associated statistical samples. In this paper, it is shown how a recently developed procedure for probabilistic learning on manifolds can serve to improve the predictability in a probabilistic framework of a scramjet simulation. The estimates of the probability density functions of the quantities of interest are improved together with estimates of the statistics of their maxima. Lastly, it is also demonstrated how the improved statistical model adds critical insight to the performance of the model.

42 ENGINEERING↗

ARM Airborne Continuous carbon dioxide measurements

The heart of the AOS CO2 Airborne Rack Mounted Analyzer System is the AOS Manifold. The AOS Manifold is a nickel coated aluminum analyzer and gas processor designed around two identical nickel-plated gas cells, one for reference gas and one for sample gas. The sample and reference cells are uniquely designed to provide optimal flushing efficiency. These cells are situated between a black-body radiation source and a photo-diode detection system. The AOS manifold also houses flow meters, pressure sensors and control valves. The exhaust from the analyzer flows into a buffer volume which allows for precise pressure control of the analyzer. The final piece of the analyzer is the demodulator board which is used to convert the DC signal generated by the analyzer into an AC response. The resulting output from the demodulator board is an averaged count of CO2 over a specified hertz cycle reported in volts and a corresponding temperature reading. The system computer is responsible for the input of commands and therefore works to control the unit functions such as flow rate, pressure, and valve control.The remainder of the system consists of compressors, reference gases, air drier, electrical cables, and the necessary connecting plumbing to provide a dry sample air stream and reference air streams to the AOS manifold.

54 ENVIRONMENTAL SCIENCES↗

ARM Airborne Continuous carbon dioxide measurements

The heart of the AOS CO2 Airborne Rack Mounted Analyzer System is the AOS Manifold. The AOS Manifold is a nickel coated aluminum analyzer and gas processor designed around two identical nickel-plated gas cells, one for reference gas and one for sample gas. The sample and reference cells are uniquely designed to provide optimal flushing efficiency. These cells are situated between a black-body radiation source and a photo-diode detection system. The AOS manifold also houses flow meters, pressure sensors and control valves. The exhaust from the analyzer flows into a buffer volume which allows for precise pressure control of the analyzer. The final piece of the analyzer is the demodulator board which is used to convert the DC signal generated by the analyzer into an AC response. The resulting output from the demodulator board is an averaged count of CO2 over a specified hertz cycle reported in volts and a corresponding temperature reading. The system computer is responsible for the input of commands and therefore works to control the unit functions such as flow rate, pressure, and valve control.The remainder of the system consists of compressors, reference gases, air drier, electrical cables, and the necessary connecting plumbing to provide a dry sample air stream and reference air streams to the AOS manifold.

54 ENVIRONMENTAL SCIENCES↗

Nonparametric maximum likelihood estimation of probability densities by penalty function methods

When it is known a priori exactly to which finite dimensional manifold the probability density function gives rise to a set of samples, the parametric maximum likelihood estimation procedure leads to poor estimates and is unstable; while the nonparametric maximum likelihood procedure is undefined. A very general theory of maximum penalized likelihood estimation which should avoid many of these difficulties is presented. It is demonstrated that each reproducing kernel Hilbert space leads, in a very natural way, to a maximum penalized likelihood estimator and that a well-known class of reproducing kernel Hilbert spaces gives polynomial splines as the nonparametric maximum penalized likelihood estimates.

Demontricher, G. F.↗

Water quality monitor

The preprototype water quality monitor (WQM) subsystem was designed based on a breadboard monitor for pH, specific conductance, and total organic carbon (TOC). The breadboard equipment demonstrated the feasibility of continuous on-line analysis of potable water for a spacecraft. The WQM subsystem incorporated these breadboard features and, in addition, measures ammonia and includes a failure detection system. The sample, reagent, and standard solutions are delivered to the WQM sensing manifold where chemical operations and measurements are performed using flow through sensors for conductance, pH, TOC, and NH3. Fault monitoring flow detection is also accomplished in this manifold assembly. The WQM is designed to operate automatically using a hardwired electronic controller. In addition, automatic shutdown is incorporated which is keyed to four flow sensors strategically located within the fluid system.

West, S.↗

Wavelength Dispersive X-ray Fluorescence Analysis of Actinides in Dissolved Nuclear Fuels

There is an urgent need for an instrument that can quickly measure the concentration of Plutonium and other Actinides mixed with Uranium in liquids containing dissolved spent fuel rods. Parallax Research, Inc. proposes to develop an x-ray spectrometer capable of measuring U, Np and Pu in dissolved nuclear fuel rod material to less than 10 ppm levels to aid in material process control for these nuclear materials. Due to system noise produced by high radioactivity, previous x-ray spectrometers were not capable of low level measurements but the system Parallax proposed has no direct path for undesired radiation to get to the detector and the detector in the proposed device is well shielded from scatter and has very low dark current. In addition, the proposed spectrometer could measure these three elements simultaneously, also measuring background positions with an energy resolution of roughly 100 eV making it possible to see a small amount of Pu that would be hidden under the tail of the U peak in energy dispersive spectrometers. Another nearly identical spectrometer could be used to target Am and Cm if necessary. The proposed spectrometer needs only a tiny sample of roughly 1 micro-liter (1 mm 3 ) and the measurement can be done with the liquid flowing in a radiation and chemical immune quartz capillary protected by a stainless steel rod making it possible to continuously monitor the liquid or to use a capillary manifold to measure other liquid streams. Unlike other methods such as mass spectroscopy where the sample must be taken to a remote facility and might take days for turn-around, the proposed measurement should take less than an hour. This spectrometer could enable near real-time measurement of U, Pu and Np in dilute dissolved spent nuclear fuel rod streams.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Nanoscale Surface Plasmonics Sensor With Nanofluidic Control

Conventional quantitative protein assays of bodily fluids typically involve multiple steps to obtain desired measurements. Such methods are not well suited for fast and accurate assay measurements in austere environments such as spaceflight and in the aftermath of disasters. Consequently, there is a need for a protein assay technology capable of routinely monitoring proteins in austere environments. For example, there is an immediate need for a urine protein assay to assess astronaut renal health during spaceflight. The disclosed nanoscale surface plasmonics sensor provides a core detection method that can be integrated to a lab-on-chip device that satisfies the unmet need for such a protein assay technology. Assays based upon combinations of nanoholes, nanorings, and nanoslits with transmission surface plasmon resonance (SPR) are used for assays requiring extreme sensitivity, and are capable of detecting specific analytes at concentrations as low as picomole to femtomole level in well-controlled environments. The device operates in a transmission mode configuration in which light is directed at one planar surface of the array, which functions as an optical aperture. The incident light induces surface plasmon light transmission from the opposite surface of the array. The presence of a target analyte is detected by changes in the spectrum of light transmitted by the array when a target analyte induces a change in the refractive index of the fluid within the nanochannels. This occurs, for example, when a target analyte binds to a receptor fixed to the walls of the nanochannels in the array. Independent fluid handling capability for individual nanoarrays on a nanofluidic chip containing a plurality of nanochannel arrays allows each array to be used to sense a different target analyte and/or for paired arrays to analyze control and test samples simultaneously in parallel. The present invention incorporates transmission mode nanoplasmonics and nanofluidics into a single, microfluidically controlled device. The device comprises one or more arrays of aligned nanochannels that are in fluid communication with inflowing and outflowing fluid handling manifolds that control the flow of fluid through the arrays. The array acts as an aperture in a plasmonic sensor. Fluid, in the form of a liquid or a gas and comprising a sample for analysis, is moved from an inlet manifold through the nanochannel array, and out through an exit manifold. The fluid may also contain a reagent used to modify the interior surfaces of the nanochannels, and/or a reagent required for the detection of an analyte.

Wei, Jianjun↗

Trajectory Optimization via Unsupervised Probabilistic Learning On Manifolds

This report investigates the use of unsupervised probabilistic learning techniques for the analysis of hypersonic trajectories. The algorithm first extracts the intrinsic structure in the data via a diffusion map approach. Using the diffusion coordinates on the graph of training samples, the probabilistic framework augments the original data with samples that are statistically consistent with the original set. The augmented samples are then used to construct conditional statistics that are ultimately assembled in a path-planing algorithm. In this framework the controls are determined stage by stage during the flight to adapt to changing mission objectives in real-time. A 3DOF model was employed to generate optimal hypersonic trajectories that comprise the training datasets. The diffusion map algorithm identfied that data resides on manifolds of much lower dimensionality compared to the high-dimensional state space that describes each trajectory. In addition to the path-planing worflow we also propose an algorithm that utilizes the diffusion map coordinates along the manifold to label and possibly remove outlier samples from the training data. This algorithm can be used to both identify edge cases for further analysis as well as to remove them from the training set to create a more robust set of samples to be used for the path-planing process.

42 ENGINEERING↗

Collocation methods for nonlinear differential equations on low-rank manifolds

We introduce new methods for integrating nonlinear differential equations on low-rank manifolds. These methods rely on interpolatory projections onto the tangent space, enabling low-rank time integration of vector fields that can be evaluated entry-wise. A key advantage of our approach is that it does not require the vector field to exhibit low-rank structure, thereby overcoming significant limitations of traditional dynamical low-rank methods based on orthogonal projection. To construct the interpolatory projectors, we develop a sparse tensor sampling algorithm based on the discrete empirical interpolation method (DEIM) that parameterizes tensor train manifolds and their tangent spaces with cross interpolation. Using these projectors, we propose two time integration schemes on low-rank tensor train manifolds. The first scheme integrates the solution at selected interpolation indices and constructs the solution with cross interpolation. The second scheme generalizes the well-known orthogonal projector-splitting integrator to interpolatory projectors. We demonstrate the proposed methods with applications to several tensor differential equations arising from the discretization of partial differential equations.

97 MATHEMATICS AND COMPUTING↗

Ethylene monitoring and control system

A system that can accurately monitor and control low concentrations of ethylene gas includes a test chamber configured to receive sample gas potentially containing an ethylene concentration and ozone, a detector configured to receive light produced during a reaction between the ethylene and ozone and to produce signals related thereto, and a computer connected to the detector to process the signals to determine therefrom a value of the concentration of ethylene in the sample gas. The supply for the system can include a four way valve configured to receive pressurized gas at one input and a test chamber. A piston is journaled in the test chamber with a drive end disposed in a drive chamber and a reaction end defining with walls of the test chamber a variable volume reaction chamber. The drive end of the piston is pneumatically connected to two ports of the four way valve to provide motive force to the piston. A manifold is connected to the variable volume reaction chamber, and is configured to receive sample gasses from at least one of a plurality of ports connectable to degreening rooms and to supply the sample gas to the reactive chamber for reaction with ozone. The apparatus can be used to monitor and control the ethylene concentration in multiple degreening rooms.

Nelson, Bruce N.↗

Ethylene monitoring and control system

A system that can accurately monitor and control low concentrations of ethylene gas includes a test chamber configured to receive sample gas potentially containing an ethylene concentration and ozone, a detector configured to receive light produced during a reaction between the ethylene and ozone and to produce signals related thereto, and a computer connected to the detector to process the signals to determine therefrom a value of the concentration of ethylene in the sample gas. The supply for the system can include a four way valve configured to receive pressurized gas at one input and a test chamber. A piston is journaled in the test chamber with a drive end disposed in a drive chamber and a reaction end defining with walls of the test chamber a variable volume reaction chamber. The drive end of the piston is pneumatically connected to two ports of the four way valve to provide motive force to the piston. A manifold is connected to the variable volume reaction chamber, and is configured to receive sample gasses from at least one of a plurality of ports connectable to degreening rooms and to supply the sample gas to the reactive chamber for reaction with ozone. The apparatus can be used to monitor and control the ethylene concentration in multiple degreening rooms.

Nelson, Bruce N.↗

Bayesian learning of orthogonal embeddings for multi-fidelity Gaussian Processes

Uncertainty propagation in complex engineering systems often poses significant computational challenges related to modeling and quantifying probability distributions of model outputs, as those emerge as the result of various sources of uncertainty that are inherent in the system under investigation. Gaussian Processes regression (GPs) is a robust meta-modeling technique that allows for fast model prediction and exploration of response surfaces. Multi-fidelity variations of GPs further leverage information from cheap and low fidelity model simulations in order to improve their predictive performance on the high fidelity model. In order to cope with the high volume of data required to train GPs in high dimensional design spaces, a common practice is to introduce latent design variables that are typically projections of the original input space to a lower dimensional subspace, and therefore substitute the problem of learning the initial high dimensional mapping, with that of training a GP on a low dimensional space. Here in this paper, we present a Bayesian approach to identify optimal transformations that map the input points to low dimensional latent variables. The \projection" mapping consists of an orthonormal matrix that is considered a priori unknown and needs to be inferred jointly with the GP parameters, conditioned on the available training data. The proposed Bayesian inference scheme relies on a two-step iterative algorithm that samples from the marginal posteriors of the GP parameters and the projection matrix respectively, both using Markov Chain Monte Carlo (MCMC) sampling. In order to take into account the orthogonality constraints imposed on the orthonormal projection matrix, a Geodesic Monte Carlo sampling algorithm is employed, that is suitable for exploiting probability measures on manifolds. We extend the proposed framework to multi-fidelity models using GPs including the scenarios of training multiple outputs together. We validate our framework on three synthetic problems with a known lower-dimensional subspace. The benefits of our proposed framework, are illustrated on the computationally challenging aerodynamic optimization of a last-stage blade for an industrial gas turbine, where we study the effect of an 85-dimensional shape parameterization of a three-dimensional airfoil on two output quantities of interest, specifically on the aerodynamic efficiency and the degree of reaction

42 ENGINEERING↗

Transitional Markov Chain Monte Carlo Sampler in UQTk

Transitional Markov Chain Monte Carlo (TMCMC) is a variant of a class of Markov Chain Monte Carlo algorithms known as tempering-based methods. In this report, the implementation of TMCMC in the Uncertainty Quantification Toolkit is investigated through the sampling of high-dimensional distributions, multi-modal distributions, and nonlinear manifolds. Furthermore, the Bayesian model evidence estimates obtained from TMCMC are tested on problems with known analytical solutions and shown to provide consistent results.

97 MATHEMATICS AND COMPUTING↗

Low gravity exothermic heating/cooling apparatus

A low gravity exothermic heating/cooling apparatus is disclosed for processing materials in space which includes an insulated casing and a sample support carried within the casing which support a sample container. An exothermic heat source includes a plurality of segments of exothermic material stacked one upon another to produce a desired temperature profile when ignited. The sample container is arranged within the core of the stacked exothermic heating material. Igniters are spaced vertically along the axis of the heating material to ignite the exothermic material at spaced points to provide total rapid burn and release of heat. To rapidly cool and quench the heat, a source of liquid carbon dixoide is provided which is conveyed through a conduit and a metering orifice into a distribution manifold where the carbon dioxide is gasified and dispersed around the exothermic heating material and the sample container via tubes for rapidly cooling the material sample.

Poorman, R. M.↗