SEARCH · Engineering Papers

Results for “generalized functions”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,243 records · Page 69

All sources NASA NTRS DOE OSTI

NASA NTRS2004

Carrier Modulation Via Waveform Probability Density Function

Beyond the classic modes of carrier modulation by varying amplitude (AM), phase (PM), or frequency (FM), we extend the modulation domain of an analog carrier signal to include a class of general modulations which are distinguished by their probability density function histogram. Separate waveform states are easily created by varying the pdf of the transmitted waveform. Individual waveform states are assignable as proxies for digital ONEs or ZEROs. At the receiver, these states are easily detected by accumulating sampled waveform statistics and performing periodic pattern matching, correlation, or statistical filtering. No fundamental natural laws are broken in the detection process. We show how a typical modulation scheme would work in the digital domain and suggest how to build an analog version. We propose that clever variations of the modulating waveform (and thus the histogram) can provide simple steganographic encoding.

Williams, Glenn L.↗

NASA NTRS2006

Carrier Modulation Via Waveform Probability Density Function

Beyond the classic modes of carrier modulation by varying amplitude (AM), phase (PM), or frequency (FM), we extend the modulation domain of an analog carrier signal to include a class of general modulations which are distinguished by their probability density function histogram. Separate waveform states are easily created by varying the pdf of the transmitted waveform. Individual waveform states are assignable as proxies for digital one's or zero's. At the receiver, these states are easily detected by accumulating sampled waveform statistics and performing periodic pattern matching, correlation, or statistical filtering. No fundamental physical laws are broken in the detection process. We show how a typical modulation scheme would work in the digital domain and suggest how to build an analog version. We propose that clever variations of the modulating waveform (and thus the histogram) can provide simple steganographic encoding.

Williams, Glenn L.↗

DOE OSTI2021

From Molecules to Solids: A vdW-DF-C09 Case Study of the Mercury Dihalides

The mercury dihalides show a remarkable diversity in the structural preferences in their minimum energy structure types, spanning molecular to strongly bound ionic solids. A challenge in the development of density functional methods for extended systems is to arrive at strategies that serve equally well such a broad range of bonding modes or structural preferences. The chemical bonding and the stabilities of mercury dihalides and the general utility and reliability of the van der Waals density functional with C09 exchange (vdW-DF-C09) in predicting or describing the energetics and structural preferences in these metal dihalides is examined. Here, we show that, in contrast with the uncorrected generalized gradient approximation of the Perdew-Burke-Erzenhoff (PBE) exchange-correlation functional, qualitative and quantitative patterns in the bonding of the mercury dihalide solids are well reproduced with vdW-DF-C09 for the full series of HgX 2 systems for X = F, Cl, Br, and I. The possible existence of a low-temperature cotunnite polymorph for HgF 2 and PbF 2 is posited.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

NASA NTRS1979

Computer analysis of shells of revolution using asymptotic results

It is suggested that asymptotic results for the behavior of thin shells can be incorporated in a general computer code for the analysis of a complex shell structure. The advantage when compared to existing finite difference or finite element codes is a substantial reduction in computational labor with the capability of working to a specified level of accuracy. A reduction in user preparation time and dependance on user judgment is also gained, since mesh spacing can be internally generated. The general theory is described in this paper, as well as the implementation in the computer code FAST 1 (Functional Algorithm for Shell Theory) for the analysis of the general axisymmetric shell structure with axisymmetric loading.

Steele, C. R.↗

DOE OSTI2023

Integration of Ag-CBRAM crossbars and Mott ReLU neurons for efficient implementation of deep neural networks in hardware

In-memory computing with emerging non-volatile memory devices (eNVMs) has shown promising results in accelerating matrix-vector multiplications. However, activation function calculations are still being implemented with general processors or large and complex neuron peripheral circuits. Here, we present the integration of Ag-based conductive bridge random access memory (Ag-CBRAM) crossbar arrays with Mott rectified linear unit (ReLU) activation neurons for scalable, energy and area-efficient hardware (HW) implementation of deep neural networks. We develop Ag-CBRAM devices that can achieve a high ON/OFF ratio and multi-level programmability. Compact and energy-efficient Mott ReLU neuron devices implementing ReLU activation function are directly connected to the columns of Ag-CBRAM crossbars to compute the output from the weighted sum current. We implement convolution filters and activations for VGG-16 using our integrated HW and demonstrate the successful generation of feature maps for CIFAR-10 images in HW. Our approach paves a new way toward building a highly compact and energy-efficient eNVMs-based in-memory computing system.

Mott insulators↗

DOE OSTI2024

Effect of XC functionals and dispersion corrections on the DFT‐computed structural and vibrational properties of SrCl 2 –NaCl and ZrF 4 –LiF

Density functional theory (DFT) calculations were performed to examine the impact of exchange–correlation (XC) functionals and van der Waals corrections (specifically the D3 method) on the structural and vibrational properties of the SrCl 2 –NaCl and ZrF 4 –LiF salt systems. Multiple XC functionals, including the local density approximation (LDA), the generalized gradient approximation using the Perdew–Burke–Ernzerhof (PBE) model, and its modified form suitable for solids (PBEsol), the dispersion-corrected PBE-D3 and PBEsol-D3, were considered. Of these functionals, LDA was found to exhibit the highest degree of error, while PBEsol and PBE-D3 displayed the least error. Underestimated lattice parameters compared with experimental values were observed to result in higher force constants, leading to an overprediction of vibrational frequencies. Conversely, an overestimation of lattice parameters was associated with lower vibrational frequencies. The methodology presented in this study yielded results that are in good agreement with experiment, irrespective of the method (finite differences vs. density functional perturbation theory) employed for calculating infrared and Raman spectra. It was further demonstrated that for alkali halides with weak Raman scattering, utilizing a supercell constructed from primitive cells better predicts Raman features than does the use of conventional cells.

Raman↗

NASA NTRS2002

Overlap Properties of Clouds Generated by a Cloud Resolving Model

In order for General Circulation Models (GCMs), one of our most important tools to predict future climate, to correctly describe the propagation of solar and thermal radiation through the cloudy atmosphere a realistic description of the vertical distribution of cloud amount is needed. Actually, one needs not only the cloud amounts at different levels of the atmosphere, but also how these cloud amounts are related, in other words, how they overlap. Currently GCMs make some idealized assumptions about cloud overlap, for example that contiguous cloud layers overlap maximally and non-contiguous cloud layers overlap in a random fashion. Since there are difficulties in obtaining the vertical profile of cloud amount from observations, the realism of the overlap assumptions made in GCMs has not been yet rigorously investigated. Recently however, cloud observations from a relatively new type of ground radar have been used to examine the vertical distribution of cloudiness. These observations suggest that the GCM overlap assumptions are dubious. Our study uses cloud fields from sophisticated models dedicated to simulate cloud formation, maintenance, and dissipation called Cloud Resolving Models . These models are generally considered capable of producing realistic three-dimensional representation of cloudiness. Using numerous cloud fields produced by such a CRM we show that the degree of overlap between cloud layers is a function of their separation distance, and is in general described by a combination of the maximum and random overlap assumption, with random overlap dominating as separation distances increase. We show that it is possible to parameterize this behavior in a way that can eventually be incorporated in GCMs. Our results seem to have a significant resemblance to the results from the radar observations despite the completely different nature of the datasets. This consistency is encouraging and will promote development of new radiative transfer codes that will estimate the radiation effects of multi-layer cloud fields more accurately.

Oreopoulos, L.↗

NASA NTRS1998

Gas-Kinetic Theory Based Flux Splitting Method for Ideal Magnetohydrodynamics

A gas-kinetic solver is developed for the ideal magnetohydrodynamics (MHD) equations. The new scheme is based on the direct splitting of the flux function of the MHD equations with the inclusion of "particle" collisions in the transport process. Consequently, the artificial dissipation in the new scheme is much reduced in comparison with the MHD Flux Vector Splitting Scheme. At the same time, the new scheme is compared with the well-developed Roe-type MHD solver. It is concluded that the kinetic MHD scheme is more robust and efficient than the Roe- type method, and the accuracy is competitive. In this paper the general principle of splitting the macroscopic flux function based on the gas-kinetic theory is presented. The flux construction strategy may shed some light on the possible modification of AUSM- and CUSP-type schemes for the compressible Euler equations, as well as to the development of new schemes for a non-strictly hyperbolic system.

Xu, Kun↗

DOE OSTI2023

QCD resummation of dijet azimuthal decorrelations in pp and pA collisions

We study the azimuthal angular decorrelations of dijet production in both proton-proton (pp) and proton-nucleus (pA) collisions. By utilizing soft-collinear effective theory, we establish the factorization and resummation formalism at the next-to-leading logarithmic accuracy for the azimuthal angular decorrelations in the back-to-back limit in pp collisions. We propose an approach where the nuclear modifications to dijet production in pA collisions are accounted for in the nuclear modified transverse momentum dependent parton distribution functions (nTMDPDFs), which contain both collinear and transverse dynamics. This approach naturally generalizes the well-established formalism related to the nuclear modified collinear parton distribution functions (nPDFs). We demonstrate strong consistency between our methodology and the CMS measurements in both pp and pA collisions, and make predictions for dijet production in the forward rapidity region in pA collisions at LHC kinematics and for mid-rapidity kinematics at sPHENIX. Throughout this paper, we focus on the application of this formalism to a simultaneous fit to both collinear and transverse momentum dependent contributions to the transverse momentum dependent distributions.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

NASA NTRS1992

Intrinsically asymmetric astrophysical jets

Previous work on the origin of self-collimated electromagnetic jets is extended to the general case where there is no reflection symmetry of the magnetic field about the equatorial plane of the disk. The axisymmetric field structure inside the disk is obtained by solving for the magnetic flux function, and the toroidal magnetic field, from the generalized thin-disk induction equation, which is derived. The asymptotic (large-z) magnetic field structure outside the disk is obtained by solving the force-free Grad-Shafranov equation semianalytically. Jet solutions are found in which the power flow is carried mainly by the Poynting flux of the electromagnetic field and the angular momentum outflow from the disk is carried by the magnetic field. The ratio of jet luminosities (top/bottom) depends directly on the degree of asymmetry of the field and can easily be much greater than unity. It is argued that the degree of field asymmetry in the disk is determined by the asymmetry of the weak galactic field fed into the disk at large distances over long periods of time.

Wang, J. C. L.↗

DOE OSTI2025

Parametric matrix models

We present a general class of machine learning algorithms called parametric matrix models. In contrast with most existing machine learning models that imitate the biology of neurons, parametric matrix models use matrix equations that emulate physical systems. Similar to how physics problems are usually solved, parametric matrix models learn the governing equations that lead to the desired outputs. Parametric matrix models can be efficiently trained from empirical data, and the equations may use algebraic, differential, or integral relations. While originally designed for scientific computing, we prove that parametric matrix models are universal function approximators that can be applied to general machine learning problems. After introducing the underlying theory, we apply parametric matrix models to a series of different challenges that show their performance for a wide range of problems. For all the challenges tested here, parametric matrix models produce accurate results within an efficient and interpretable computational framework that allows for input feature extrapolation.

Computational science↗

DOE OSTI2025

Fluorogenic Biosensing with Tunable Polydiacetylene Vesicles

Polydiacetylenes (PDAs) are conjugated polymers that are well known for their colorimetric transition from blue to red with the application of energetic stimulus. Sensing platforms based on polymerized diacetylene surfactant vesicles and other structures have been widely demonstrated for various colorimetric biosensing applications. Although less studied and utilized, the transition also results in a change from a non-fluorescent to a highly fluorescent state, making polydiacetylenes useful for both colorimetric and fluorogenic sensing applications. Here, we focus on the characterization and optimization of polydiacetylene vesicles to tune their sensitivity for fluorogenic sensing applications. Particularly, we look at how the structure of the diacetylene (DA) hydrocarbon tail and headgroup affect the self-assembled vesicle size and stability, polymerization kinetics, and the fluorogenic, blue to red phase transition. Longer DA acyl tails generally resulted in smaller and more stable vesicles. The polymerization kinetics and the blue to red transition were a function of both the DA acyl tail length and structure of the headgroup. Decreasing the acyl tail length generally led to vesicles that were more sensitive to energetic stimuli. Headgroup modifications had different effects depending on the structure of the headgroup. Ethanolamine headgroups resulted in vesicles with potentially increased stimuli responsivity. The lower energy stimulus to induce the chromatic transition was attributed to an increase in headgroup hydrogen bonding and polymer backbone strain. Boronic-acid headgroup functionalization led to vesicles that were generally unstable, only weakly polymerized, and unable to fully transform to the red phase due to strong polar, aromatic headgroup interactions. This work presents the design of PDA vesicles in the context of biosensing platforms and includes a discussion of the past, present, and future of PDA biosensing.

Chemistry↗

DOE OSTI2021

Deciphering enhancer sequence using thermodynamics-based models and convolutional neural networks

Abstract Deciphering the sequence-function relationship encoded in enhancers holds the key to interpreting non-coding variants and understanding mechanisms of transcriptomic variation. Several quantitative models exist for predicting enhancer function and underlying mechanisms; however, there has been no systematic comparison of these models characterizing their relative strengths and shortcomings. Here, we interrogated a rich data set of neuroectodermal enhancers in Drosophila, representing cis- and trans- sources of expression variation, with a suite of biophysical and machine learning models. We performed rigorous comparisons of thermodynamics-based models implementing different mechanisms of activation, repression and cooperativity. Moreover, we developed a convolutional neural network (CNN) model, called CoNSEPT, that learns enhancer ‘grammar’ in an unbiased manner. CoNSEPT is the first general-purpose CNN tool for predicting enhancer function in varying conditions, such as different cell types and experimental conditions, and we show that such complex models can suggest interpretable mechanisms. We found model-based evidence for mechanisms previously established for the studied system, including cooperative activation and short-range repression. The data also favored one hypothesized activation mechanism over another and suggested an intriguing role for a direct, distance-independent repression mechanism. Our modeling shows that while fundamentally different models can yield similar fits to data, they vary in their utility for mechanistic inference. CoNSEPT is freely available at: https://github.com/PayamDiba/CoNSEPT.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2021

PayamDiba/CoNSEPT

Deciphering the sequence-function relationship encoded in enhancers holds the key to interpreting non-coding variants and understanding mechanisms of transcriptomic variation. Several quantitative models exist for predicting enhancer function and underlying mechanisms; however, there has been no systematic comparison of these models characterizing their relative strengths and shortcomings. Here, we interrogated a rich data set of neuroectodermal enhancers in Drosophila, representing cis- and trans- sources of expression variation, with a suite of biophysical and machine learning models. We performed rigorous comparisons of thermodynamics-based models implementing different mechanisms of activation, repression and cooperativity. Moreover, we developed a convolutional neural network (CNN) model, called CoNSEPT, that learns enhancer ‘grammar’ in an unbiased manner. CoNSEPT is the first general-purpose CNN tool for predicting enhancer function in varying conditions, such as different cell types and experimental conditions, and we show that such complex models can suggest interpretable mechanisms. We found model-based evidence for mechanisms previously established for the studied system, including cooperative activation and short-range repression. The data also favored one hypothesized activation mechanism over another and suggested an intriguing role for a direct, distance-independent repression mechanism. Our modeling shows that while fundamentally different models can yield similar fits to data, they vary in their utility for mechanistic inference.

Dibaeinia, Payam↗

DOE OSTI2021

A selection function toolbox for subsets of astronomical catalogues

ABSTRACT Large catalogues are ubiquitous throughout astronomy, but most scientific analyses are carried out on smaller samples selected from these catalogues by chosen cuts on catalogued quantities. The selection function of that scientific sample – the probability that a star in the catalogue will satisfy these cuts and so make it into the sample – is thus unique to each scientific analysis. We have created a general framework that can flexibly estimate the selection function of a sample drawn from a catalogue in terms of position, magnitude and colour. Our method is unique in using the binomial likelihood and accounting for correlations in the selection function across position, magnitude, and colour using Gaussian processes and spherical harmonics. We have created a new open-source python package selectionfunctiontoolbox that implements this framework and used it to make three different estimates of the APOGEE DR16 red giant sample selection function, as a subset of 2MASS, with each estimate using an increasing amount of technical knowledge of the APOGEE targeting. In a companion paper, we applied our methodology to derive estimates of the astrometric and spectroscopic selection functions of Gaia EDR3. Our framework will make it trivial for astrophysicists to estimate the selection function that they should be using with the custom sample of stars that they have chosen to answer their scientific question.

79 ASTRONOMY AND ASTROPHYSICS↗

NASA NTRS1990

Mappings and accuracy for Chebyshev pseudo-spectral approximations

The effect of mappings on the approximation, by Chebyshev collocation, of functions which exhibit localized regions of rapid variation is studied. A general strategy is introduced whereby mappings are adaptively constructed which map specified classes of rapidly varying functions into low order polynomials which can be accurately approximated by Chebyshev polynomial expansions. A particular family of mappings constructed in this way is tested on a variety of rapidly varying functions similar to those occurring in approximations. It is shown that the mapped function can be approximated much more accurately by Chebyshev polynomial approximations than in physical space or where mappings constructed from other strategies are employed.

Bayliss, Alvin↗

NASA NTRS1992

Mappings and accuracy for Chebyshev pseudo-spectral approximations

Bayliss, Alvin↗

DOE OSTI2022

MICCO: An Enhanced Multi-GPU Scheduling Framework for Many-Body Correlation Functions

Calculation of many-body correlation functions is one of the critical kernels utilized in many scientific computing areas, especially in Lattice Quantum Chromodynamics (Lattice QCD). It is formalized as a sum of a large number of contraction terms each of which can be represented by a graph consisting of vertices describing quarks inside a hadron node and edges designating quark propagations at specific time intervals. Due to its computation- and memory-intensive nature, real-world physics systems (e.g., multi-meson or multi-baryon systems) explored by Lattice QCD prefer to leverage multi-GPUs. Different from general graph processing, many-body correlation function calculations show two specific features: a large number of computation-/data-intensive kernels and frequently repeated appearances of original and intermediate data. The former results in expensive memory operations such as tensor movements and evictions. The latter offers data reuse opportunities to mitigate the data-intensive nature of many-body correlation function calculations. However, existing graph-based multi-GPU schedulers cannot capture these data-centric features, thus resulting in a sub-optimal performance for many-body correlation function calculations. To address this issue, this paper presents a multi-GPU scheduling framework, MICCO, to accelerate contractions for correlation functions particularly by taking the data dimension (e.g., data reuse and data eviction) into account. This work first performs a comprehensive study on the interplay of data reuse and load balance, and designs two new concepts: local reuse pattern and reuse bound to study the opportunity of achieving the optimal trade-off between them. Based on this study, MICCO proposes a heuristic scheduling algorithm and a machine-learning-based regression model to generate the optimal setting of reuse bounds. Specifically, MICCO is integrated into a real-world Lattice QCD system, Redstar, for the first time running on multiple GPUs. The evaluation demonstrates MICCO outperforms other state-of-art works, achieving up to 2.25× speedup in synthesized datasets, and 1.49× speedup in real-world correlation functions.

Wang, Qihan↗