SEARCH · Search NASA
Results for “NN”
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.
Physics-informed graph neural nets (pigNNS) A unification of NN architectures with mimetic PDE discretization.
Abstract not provided.
Physics-informed graph neural netsA unification of NN architectures with mimetic PDE discretization.
Abstract not provided.
Fragmentation of jets containing a prompt J$/\psi$ meson in PbPb and pp collisions at $\sqrt{s_\mathrm{NN}} =$ 5.02 TeV
Jets containing a prompt J$/\psi$ meson are studied in lead-lead collisions at a nucleon-nucleon center-of-mass energy of 5.02 TeV, using the CMS detector at the LHC. Jets are selected to be in the transverse momentum range of 30 $\lt p_\mathrm{T} \lt$ 40 GeV. The J$/\psi$ yield in these jets is evaluated as a function of the jet fragmentation variable $z$, the ratio of the J$/\psi$$p_\mathrm{T} $ to the jet $p_\mathrm{T}$. The nuclear modification factor, $R_\mathrm{AA}$, is then derived by comparing the yield in lead-lead collisions to the corresponding expectation based on proton-proton data, at the same nucleon-nucleon center-of-mass energy. The suppression of the J$/\psi$ yield shows a dependence on $z$, indicating that the interaction of the J$/\psi$ with the quark-gluon plasma formed in heavy ion collisions depends on the fragmentation that gives rise to the J$/\psi$ meson.
Femtosecond 1030-nn Mirror Thin Film Damage Competition
Explore the source record for details and available documents.
Luminosity measurement for lead-lead collisions at $\sqrt{s_{\mathrm{NN}}}$ = 5.02 TeV in 2015 and 2018 at CMS
Measurements of the luminosity delivered to the CMS experiment during the lead-lead data-taking periods in 2015 and 2018 are presented for the first time. The collisions were recorded at a nucleon-nucleon center-of-mass energy of 5.02 TeV; the 2018 data sample is three times larger than the 2015 data sample. Three subdetectors are used: the pixel luminosity telescope, the forward hadron calorimeters, and the fast beam conditions monitor. The absolute luminosity calibration is determined using the van der Meer technique that relies on transverse beam separation scans. The dominant sources of uncertainty are the transverse factorizability of the bunch density profiles and, in 2015, the difference between the results obtained using various detectors. The total uncertainty in the integrated luminosity, including the stability of the calibrated subdetector response over time, amounts to 3.0% for 2015, 1.7% for 2018, and 1.6% for the combined data sample.
Photoinduced Magnetic Exchange-Jump Promotes Ground State Biradical Electron Spin Polarization
Photoinduced electron spin polarization (ESP) is reported in the electronic ground states of three Pt(II) complexes that are comprised of two S = ½ nitronyl nitroxide (NN) radicals attached through different length para-phenylethynyl bridges to the 3,6 positions of a catecholate (CAT, donor) and 4,4’-di-tert-butyl-2,2’-bipyridine (bpy, acceptor). Complexes 1-3 have from 17 to 41 bonds separating NN radicals, and display cw-EPR spectra consistent with |J NN-NN | >> |a N |, |J NN-NN | ≥ |a N |, and |J NN-NN | < |a N |, respectively, where J NN-NN is the magnetic exchange coupling between NN radicals in the electronic ground state, and a N is the isotropic 14 N hyperfine coupling constant. Light-induced transient EPR spectra characterized as enhanced ground-state absorption were observed for all three complexes using 532 nm pulsed laser excitation into the ligand-to-ligand charge transfer (LL’CT) band of the (CAT)Pt(bpy) chromophore. The magnitude of the observed ESP increases in the order 1 < 2 < 3 and is inversely correlated with the magnitude of ground-state J NN-NN . In addition to the experimental observation net absorptive polarization in 1-3, light excitation also produces multiplet polarization in 2. Since the weak dipolar coupling leads to strong spectral overlap of the absorptive and emissive components, the multiplet polarization is not observed in 1 and 3 and is very weak in 2. Here, the ability to spin polarize multiple radical spins with a single photon is anticipated to advance new photoinduced multi qubit/qudit ESP protocols for quantum information science applications.
Neural Net Safety Monitor Design
The National Aeronautics and Space Administration (NASA) at the Dryden Flight Research Center (DFRC) has been conducting flight-test research using an F-15 aircraft (figure 1). This aircraft has been specially modified to interface a neural net (NN) controller as part of a single-string Airborne Research Test System (ARTS) computer with the existing quad-redundant flight control system (FCC) shown in figure 2. The NN commands are passed to FCC channels 2 and 4 and are cross channel data linked (CCDL) to the other computers as shown. Numerous types of fault-detection monitors exist in the FCC when the NN mode is engaged; these monitors would cause an automatic disengagement of the NN in the event of a triggering fault. Unfortunately, these monitors still may not prevent a possible NN hard-over command from coming through to the control laws. Therefore, an additional and unique safety monitor was designed for a single-string source that allows authority at maximum actuator rates but protects the pilot and structural loads against excessive g-limits in the case of a NN hard-over command input. This additional monitor resides in the FCCs and is executed before the control laws are computed. This presentation describes a floating limiter (FL) concept1 that was developed and successfully test-flown for this program (figure 3). The FL computes the rate of change of the NN commands that are input to the FCC from the ARTS. A window is created with upper and lower boundaries, which is constantly floating and trying to stay centered as the NN command rates are changing. The limiter works by only allowing the window to move at a much slower rate than those of the NN commands. Anywhere within the window, however, full rates are allowed. If a rate persists in one direction, it will eventually hit the boundary and be rate-limited to the floating limiter rate. When this happens, a persistent counter begins and after a limit is reached, a NN disengage command is generated. The tunable metrics for the FL are (1) window size, (2) drift rate, and (3) persistence counter. Ultimate range limits are also included in case the NN command should drift slowly to a limit value that would cause the FL to be defeated. The FL has proven to work as intended. Both high-g transients and excessive structural loads are controlled with NN hard-over commands. This presentation discusses the FL design features and presents test cases. Simulation runs are included to illustrate the dramatic improvement made to the control of NN hard-over effects. A mission control room display from a flight playback is presented to illustrate the neural net fault display representation. The FL is very adaptable to various requirements and is independent of flight condition. It should be considered as a cost-effective safety monitor to control single-string inputs in general.
Implementation of an Adaptive Controller System from Concept to Flight Test
The National Aeronautics and Space Administration (NASA) at the Dryden Flight Research Center (DFRC) has been conducting flight-test research using an F-15 aircraft (figure 1). This aircraft has been specially modified to interface a neural net (NN) controller as part of a single-string Airborne Research Test System (ARTS) computer with the existing quad-redundant flight control system (FCC) shown in figure 2. The NN commands are passed to FCC channels 2 and 4 and are cross channel data linked (CCDL) to the other computers as shown. Numerous types of fault-detection monitors exist in the FCC when the NN mode is engaged; these monitors would cause an automatic disengagement of the NN in the event of a triggering fault. Unfortunately, these monitors still may not prevent a possible NN hard-over command from coming through to the control laws. Therefore, an additional and unique safety monitor was designed for a single-string source that allows authority at maximum actuator rates but protects the pilot and structural loads against excessive g-limits in the case of a NN hard-over command input. This additional monitor resides in the FCCs and is executed before the control laws are computed. This presentation describes a "floating limiter" (FL) concept that was developed and successfully test-flown for this program (figure 3). The FL computes the rate of change of the NN commands that are input to the FCC from the ARTS. A window is created with upper and lower boundaries, which is constantly "floating" and trying to stay centered as the NN command rates are changing. The limiter works by only allowing the window to move at a much slower rate than those of the NN commands. Anywhere within the window, however, full rates are allowed. If a rate persists in one direction, it will eventually "hit" the boundary and be rate-limited to the floating limiter rate. When this happens, a persistent counter begins and after a limit is reached, a NN disengage command is generated. The tunable metrics for the FL are (1) window size, (2) drift rate, and (3) persistence counter. Ultimate range limits are also included in case the NN command should drift slowly to a limit value that would cause the FL to be defeated. The FL has proven to work as intended. Both high-g transients and excessive structural loads are controlled with NN hard-over commands. This presentation discusses the FL design features and presents test cases. Simulation runs are included to illustrate the dramatic improvement made to the control of NN hard-over effects. A mission control room display from a flight playback is presented to illustrate the neural net fault display representation. The FL is very adaptable to various requirements and is independent of flight condition. It should be considered as a cost-effective safety monitor to control single-string inputs in general.
WNN 92; Proceedings of the 3rd Workshop on Neural Networks: Academic/Industrial/NASA/Defense, Auburn Univ., AL, Feb. 10-12, 1992 and South Shore Harbour, TX, Nov. 4-6, 1992
The present conference discusses such neural networks (NN) related topics as their current development status, NN architectures, NN learning rules, NN optimization methods, NN temporal models, NN control methods, NN pattern recognition systems and applications, biological and biomedical applications of NNs, VLSI design techniques for NNs, NN systems simulation, fuzzy logic, and genetic algorithms. Attention is given to missileborne integrated NNs, adaptive-mixture NNs, implementable learning rules, an NN simulator for travelling salesman problem solutions, similarity-based forecasting, NN control of hypersonic aircraft takeoff, NN control of the Space Shuttle Arm, an adaptive NN robot manipulator controller, a synthetic approach to digital filtering, NNs for speech analysis, adaptive spline networks, an anticipatory fuzzy logic controller, and encoding operations for fuzzy associative memories.
Origin of Ferromagnetic Exchange Coupling in Donor–Acceptor Biradical Analogues of Charge-Separated Excited States
A new donor–acceptor biradical complex, Tp Cum,Me Zn(SQ-VD) (Tp Cum,Me Zn + = zinc(II) hydro-tris(3-cumenyl-5-methylpyrazolyl)borate complex cation; SQ = orthosemiquinone; VD = oxoverdazyl), which is a ground-state analogue of a charge-separated excited state, has been synthesized and structurally characterized. Here, the magnetic exchange interaction between the S = 1/2 SQ and the S = 1/2 VD within the SQ-VD biradical ligand is observed to be ferromagnetic, with $J_{SQ-VD}$ = +77 cm –1 (H = -2$J_{SQ-VD}\bar{S}_{SQ}·\bar{S}_{VD}$) determined from an analysis of the variable-temperature magnetic susceptibility data. The pairwise biradical exchange interaction in Tp Cum,Me Zn(SQ-VD) can be compared with that of the related donor–acceptor biradical complex Tp Cum,Me Zn(SQ-NN) (NN = nitronyl nitroxide, S = 1/2), where $J_{SQ-NN}$ ≅ +550 cm –1 . This represents a dramatic reduction in the biradical exchange by a factor of ~7, despite the isolobal nature of the VD and NN acceptor radical SOMOs. Computations assessing the magnitude of the exchange were performed using a broken-symmetry density functional theory (DFT) approach. These computations are in good agreement with those computed at the CASSCF NEVPT2 level, which also reveals an S = 1 triplet ground state as observed in the magnetic susceptibility measurements. A combination of electronic absorption spectroscopy and CASSCF computations has been used to elucidate the electronic origin of the large difference in the magnitude of the biradical exchange coupling between Tp Cum,Me Zn(SQ-VD) and Tp Cum,Me Zn(SQ-NN). A Valence Bond Configuration Interaction (VBCI) model was previously employed to highlight the importance of mixing an SQ SOMO → NNLUMO charge transfer configuration into the electronic ground state to facilitate the stabilization of the high-spin triplet (S = 1) ground state in Tp Cum,Me Zn(SQ-NN). Here, CASSCF computations confirm the importance of mixing the pendant radical (e.g., VD, NN) LUMO (VD LUMO and NN LUMO ) with the SOMO of the SQ radical (SQ SOMO ) for stabilizing the triplet, in addition to spin polarization and charge transfer contributions to the exchange. An important electronic structure difference between Tp Cum,Me Zn(SQ-VD) and Tp Cum,Me Zn(SQ-NN), which leads to their different exchange couplings, is the reduced admixture of excited states that promote ferromagnetic exchange into the Tp Cum,Me Zn(SQ-VD) ground state, and the intrinsically weaker mixing between the VD LUMO and the SQ SOMO compared to that observed for Tp Cum,Me Zn(SQ-NN), where this orbital mixing is significant. The results of this comparative study contribute to a greater understanding of biradical exchange interactions, which are important to our understanding of excited-state singlet–triplet energy gaps, electron delocalization, and the generation of electron spin polarization in both the ground and excited states of (bpy)Pt(CAT-radical) complexes.
Neural network uncertainty assessment using Bayesian statistics: a remote sensing application
Neural network (NN) techniques have proved successful for many regression problems, in particular for remote sensing; however, uncertainty estimates are rarely provided. In this article, a Bayesian technique to evaluate uncertainties of the NN parameters (i.e., synaptic weights) is first presented. In contrast to more traditional approaches based on point estimation of the NN weights, we assess uncertainties on such estimates to monitor the robustness of the NN model. These theoretical developments are illustrated by applying them to the problem of retrieving surface skin temperature, microwave surface emissivities, and integrated water vapor content from a combined analysis of satellite microwave and infrared observations over land. The weight uncertainty estimates are then used to compute analytically the uncertainties in the network outputs (i.e., error bars and correlation structure of these errors). Such quantities are very important for evaluating any application of an NN model. The uncertainties on the NN Jacobians are then considered in the third part of this article. Used for regression fitting, NN models can be used effectively to represent highly nonlinear, multivariate functions. In this situation, most emphasis is put on estimating the output errors, but almost no attention has been given to errors associated with the internal structure of the regression model. The complex structure of dependency inside the NN is the essence of the model, and assessing its quality, coherency, and physical character makes all the difference between a blackbox model with small output errors and a reliable, robust, and physically coherent model. Such dependency structures are described to the first order by the NN Jacobians: they indicate the sensitivity of one output with respect to the inputs of the model for given input data. We use a Monte Carlo integration procedure to estimate the robustness of the NN Jacobians. A regularization strategy based on principal component analysis is proposed to suppress the multicollinearities in order to make these Jacobians robust and physically meaningful.
Active Learning A Neural Network Model For Gold Clusters & Bulk From Sparse First Principles Training Data
Small metal clusters are of fundamental scientific interest and of tremendous significance in catalysis. These nanoscale clusters display diverse geometries and structural motifs depending on the cluster size; a knowledge of this size-dependent structural motifs and their dynamical evolution has been of longstanding interest. Given the high computational cost of first-principles calculations, molecular modeling and atomistic simulations such as molecular dynamics (MD) has proven to be an important complementary tool to aid this understanding. Classical MD typically employ predefined functional forms which limits their ability to capture such complex size-dependent structural and dynamical transformation. Neural Network (NN) based potentials represent flexible alternatives and in principle, well-trained NN potentials can provide high level of flexibility, transferability and accuracy on-par with the reference model used for training. A major challenge, however, is that NN models are interpolative and requires large quantities (similar to 10 4 or greater) of training data to ensure that the model adequately samples the energy landscape both near and far-from-equilibrium. A highly desirable goal is minimize the number of training data, especially if the underlying reference model is first-principles based and hence expensive. In this work, we introduce an active learning (AL) scheme that trains a NN model on-the-fly with minimal amount of first-principles based training data. Our AL workflow is initiated with a sparse training dataset (similar to 1 to 5 data points) and is updated on-the-fly via a Nested Ensemble Monte Carlo scheme that iteratively queries the energy landscape in regions of failure and updates the training pool to improve the network performance. Using a representative system of gold clusters, we demonstrate that our AL workflow can train a NN with similar to 500 total reference calculations. Using an extensive DFT test set of similar to 1100 configurations, we show that our AL-NN is able to accurately predict both the DFT energies and the forces for clusters of a myriad of different sizes. Our NN predictions are within 30 meV/atom and 40 meV/angstrom of the reference DFT calculations. Moreover, our AL-NN model also adequately captures the various size-dependent structural and dynamical properties of gold clusters in excellent agreement with DFT calculations and available experiments. We finally show that our AL-NN model also captures bulk properties reasonably well, even though they were not included in the training data.
Exploring the unknown Lambda-neutron interaction
No published \Lambda Λ n scattering data exist. A relativistic heavy-ion experiment has suggested that a \Lambda Λ nn bound state was seen. However, several theoretical analyses have cast serious doubt on the bound-state assertion. Nevertheless, there could exist a three-body \Lambda Λ nn resonance. Such a resonance could be used to constrain the \Lambda Λ n interaction. We discuss \Lambda Λ nn calculations using nn and \Lambda Λ n pairwise interactions of rank-one, separable form that fit effective range parameters of the nn system and those hypothesized for the as yet unobserved \Lambda Λ n system based upon four different \Lambda Λ N potentials. The use of rank-one separable potentials allows one to analytically continue the \Lambda Λ nn Faddeev equations onto the second complex energy plane in search of resonance poles, by examining the eigenvalue spectrum of the kernel of the Faddeev equations. Although each of the potential models predicts a \Lambda Λ nn sub-threshold resonance pole, scaling of the \Lambda Λ n interaction by as little as \sim ∼ 5% does produce a physical resonance. This suggests that one may use photo-(electro-)production of the \Lambda Λ nn system from tritium as a tool to examine the strength of the \Lambda Λ n interaction.
Estimating Surface Soil Moisture from SMAP Observations Using a Neural Network Technique
A Neural Network (NN) algorithm was developed to estimate global surface soil moisture for April 2015 to March 2017 with a 2-3 day repeat frequency using passive microwave observations from the Soil Moisture Active Passive (SMAP) satellite, surface soil temperatures from the NASA Goddard Earth Observing System Model version 5 (GEOS-5) land modeling system, and Moderate Resolution Imaging Spectroradiometer-based vegetation water content. The NN was trained on GEOS-5 soil moisture target data, making the NN estimates consistent with the GEOS-5 climatology, such that they may ultimately be assimilated into this model without further bias correction. Evaluated against in situ soil moisture measurements, the average unbiased root mean square error (ubRMSE), correlation and anomaly correlation of the NN retrievals were 0.037 m(exp. 3)m(exp. -3), 0.70 and 0.66, respectively, against SMAP core validation site measurements and 0.026 m(exp. 3)m(exp. -3), 0.58 and 0.48, respectively, against International Soil Moisture Network (ISMN) measurements. At the core validation sites, the NN retrievals have a significantly higher skill than the GEOS-5 model estimates and a slightly lower correlation skill than the SMAP Level-2 Passive (L2P) product. The feasibility of the NN method was reflected by a lower ubRMSE compared to the L2P retrievals as well as a higher skill when ancillary parameters in physically-based retrievals were uncertain. Against ISMN measurements, the skill of the two retrieval products was more comparable. A triple collocation analysis against Advanced Microwave Scanning Radiometer 2 (AMSR2) and Advanced Scatterometer (ASCAT) soil moisture retrievals showed that the NN and L2P retrieval errors have a similar spatial distribution, but the NN retrieval errors are generally lower in densely vegetated regions and transition zones.
Low-level liquid cloud properties during ORACLES retrieved using airborne polarimetric measurements and a neural network algorithm
In this study we developed a neural network (NN) that can be used to retrieve cloud microphysical properties from multiangular and multispectral polarimetric remote sensing observations. This effort builds upon our previous work, which explored the sensitivity of neural network input, architecture, and other design requirements for this type of remote sensing problem. In particular this work introduces a framework for appropriately weighting total and polarized reflectances, which have vastly different magnitudes and measurement uncertainties. The NN is trained using an artificial training set and applied to research scanning polarimeter (RSP) data obtained during the ORACLES field campaign (ObseRvations of Aerosols above CLouds and their intEractionS). The polarimetric RSP observations are unique in that they observe the same cloud from a very large number of angles within a variety of spectral bands, resulting in a large dataset that can be explored rapidly with a NN approach. The usefulness of applying a NN to a dataset such as this one stems from the possibility of rapidly obtaining a retrieval that could be subsequently applied as a first guess for slower but more rigorous physical-based retrieval algorithms. This approach could be particularly advantageous for more complicated atmospheric retrievals – such as when an aerosol layer lies above clouds like in ORACLES. For RSP observations obtained during ORACLES 2016, comparisons between the NN and standard parametric polarimetric (PP) cloud retrieval give reasonable results for droplet effective radius (r(e): R=0.756, RMSE=1.74 µm) and cloud optical thickness (τ: R=0.950, RMSE=1.82). This level of statistical agreement is shown to be similar to comparisons between the two most well-established cloud retrievals, namely, the polarimetric and the bispectral total reflectance cloud retrievals. The NN retrievals from the ORACLES 2017 dataset result in retrievals of r(e) (R=0.54, RMSE=4.77 µm) and τ (R=0.785, RMSE=5.61) that behave much more poorly. In particular we found that our NN retrieval approach does not perform well for thin (τ<3), inhomogeneous, or broken clouds. We also found that correction for above-cloud atmospheric absorption improved the NN retrievals moderately – but retrievals without this correction still behaved similarly to existing cloud retrievals with a slight systematic offset.
A neural network‐enhanced reproducing kernel particle method for modeling strain localization
Abstract Modeling the localized intensive deformation in a damaged solid requires highly refined discretization for accurate prediction, which significantly increases the computational cost. Although adaptive model refinement can be employed for enhanced effectiveness, it is cumbersome for the traditional mesh‐based methods to perform while modeling the evolving localizations. In this work, neural network‐enhanced reproducing kernel particle method (NN‐RKPM) is proposed, where the location, orientation, and shape of the solution transition near a localization is automatically captured by the NN approximation via a block‐level neural network (NN) optimization. The weights and biases in the blocked parameterization network control the location and orientation of the localization. The designed basic four‐kernel NN block is capable of capturing a triple junction or a quadruple junction topological pattern, while more complicated localization topological patters are captured by the superposition of multiple four‐kernel NN blocks. The standard RK approximation is then utilized to approximate the smooth part of the solution, which permits a much coarser discretization than the high‐resolution discretization needed to capture sharp solution transitions with the conventional methods. A regularization of the NN approximation is additionally introduced for discretization‐independent material responses. The effectiveness of the proposed NN‐RKPM is verified by a series of numerical verifications.
Lightweight and effective tensor sensitivity for atomistic neural networks
Atomistic machine learning focuses on the creation of models that obey fundamental symmetries of atomistic configurations, such as permutation, translation, and rotation invariances. In many of these schemes, translation and rotation invariance are achieved by building on scalar invariants, e.g., distances between atom pairs. There is growing interest in molecular representations that work internally with higher rank rotational tensors, e.g., vector displacements between atoms, and tensor products thereof. Here, we present a framework for extending the Hierarchically Interacting Particle Neural Network (HIP-NN) with Tensor Sensitivity information (HIP-NN-TS) from each local atomic environment. Crucially, the method employs a weight tying strategy that allows direct incorporation of many-body information while adding very few model parameters. We show that HIP-NN-TS is more accurate than HIP-NN, with negligible increase in parameter count, for several datasets and network sizes. As the dataset becomes more complex, tensor sensitivities provide greater improvements to model accuracy. In particular, HIP-NN-TS achieves a record mean absolute error of 0.927 $\frac{\text{kcal}}{\text{mol}}$ for conformational energy variation on the challenging COMP6 benchmark, which includes a broad set of organic molecules. We also compare the computational performance of HIP-NN-TS to HIP-NN and other models in the literature.