Search NASA⌕ Search

SEARCH · Search NASA

Results for “synchronized 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 199 records · Page 11

Influence of Temperature on Sr{sub 0.35}La{sub 0.40}Ca{sub .25}Fe{sub 11.6}Co{sub 0.4}O{sub 19} Hexagonal Ferrites Against Structural, Morphological and Magnetic Properties Prepared by Conventional Ceramic Reaction Methodology

Conventional Ceramic Reaction Methodology has been utilized for the preparation of M-type hexagonal ferrites of Sr{sub 0.35}La{sub 0.40}Ca{sub 0.25}Fe{sub 11.6}Co{sub 0.4}O{sub 19} magnets. X-ray diffraction has been performed for phase determination of the magnetic powders. At calcination temperatures ranging from 1220 to 1330 {sup ∘}C, the synchronize compost of the magnetic powders comprised of M-type hexagonal ferrites together with a small amount of impurity phase such as α-Fe{sub 2}O{sub 3} were co-existed. At calcination temperature above 1270 {sup ∘}C, single-phase M-type hexaferrites can be obtained. The particles seemed to be in a hexagonal plate-like mould which spread-out homogeneously were examined by field emission scanning electron microscopy. The magnetic edibility of the magnetic powders was intended by permanent magnetic measurement test equipment vibrating sample magnetometer. Magnetic properties, such as specific saturation magnetization, coercivity and remanence were temperature-dependent. However, with the increase of sintering temperature, i.e. above 1270 {sup ∘}C, saturation magnetization was increased while the coercivity showed reversed order.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Synchronous digitization for high dynamic range lock-in amplification in beam-scanning microscopy

Digital lock-in amplification (LIA) with synchronous digitization (SD) is shown to provide significant signal to noise (S/N) and linear dynamic range advantages in beam-scanning microscopy measurements using pulsed laser sources. Direct comparisons between SD-LIA and conventional LIA in homodyne second harmonic generation measurements resulted in S/N enhancements consistent with theoretical models. SD-LIA provided notably larger S/N enhancements in the limit of low light intensities, through the smooth transition between photon counting and signal averaging developed in previous work. Rapid beam scanning instrumentation with up to video rate acquisition speeds minimized photo-induced sample damage. The corresponding increased allowance for higher laser power without sample damage is advantageous for increasing the observed signal content.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Plasma membrane characterization, by scanning electron microscopy, of multipotent myoblasts-derived populations sorted using dielectrophoresis

Highlights: •Dielectrophoretic separation/sorting of multipotent cells. •Plasma membrane microvilli structure of C2C12 and fibroblasts by SEM microscopy. •Cell cycle determination by Ki-67 in DEP-sorted cells. •Plasma membrane differences responsible for changes in membrane capacitance. -- Abstract: Multipotent progenitor cells have shown promise for use in biomedical applications and regenerative medicine. The implementation of such cells for clinical application requires a synchronized, phenotypically and/or genotypically, homogenous cell population. Here we have demonstrated the implementation of a biological tag-free dielectrophoretic device used for discrimination of multipotent myoblastic C2C12 model. The multipotent capabilities in differentiation, for these cells, diminishes with higher passage number, so for cultures above 70 passages only a small percentage of cells is able to differentiate into terminal myotubes. In this work we demonstrated that we could recover, above 96% purity, specific cell types from a mixed population of cells at high passage number without any biological tag using dielectrophoresis. The purity of the samples was confirmed by cytometric analysis using the cell specific marker embryonic myosin. To further investigate the dielectric properties of the cell plasma membrane we co-culture C2C12 with similar size, when in suspension, GFP-positive fibroblast as feeder layer. The level of separation between the cell types was above 98% purity which was confirmed by flow cytometry. These levels of separation are assumed to account for cell size and for the plasma membrane morphological differences between C2C12 and fibroblast unrelated to the stages of the cell cycle which was assessed by immunofluorescence staining. Plasma membrane conformational differences were further confirmed by scanning electron microscopy.

60 APPLIED LIFE SCIENCES↗

Constraining the oblateness of Kepler planets

We use Kepler short-cadence light curves to constrain the oblateness of planet candidates in the Kepler sample. The transits of rapidly rotating planets that are deformed in shape will lead to distortions in the ingress and egress of their light curves. We report the first tentative detection of an oblate planet outside the solar system, measuring an oblateness of 0.22{sub −0.11}{sup +0.11} for the 18 M{sub J} mass brown dwarf Kepler 39b (KOI 423.01). We also provide constraints on the oblateness of the planets (candidates) HAT-P-7b, KOI 686.01, and KOI 197.01 to be <0.067, <0.251, and <0.186, respectively. Using the Q' values from Jupiter and Saturn, we expect tidal synchronization for the spins of HAT-P-7b, KOI 686.01, and KOI 197.01, and for their rotational oblateness signatures to be undetectable in the current data. The potentially large oblateness of KOI 423.01 (Kepler 39b) suggests that the Q' value of the brown dwarf needs to be two orders of magnitude larger than that of the solar system gas giants to avoid being tidally spun down.

79 ASTRONOMY AND ASTROPHYSICS↗

The Asteroid Rotation Period Survey Using the China Near-Earth Object Survey Telescope (CNEOST)

We initiated the bilateral collaboration between the Lulin Observatory and the Purple Mountain Observatory to collect asteroid lightcurves using the Chinese Near-Earth Object Survey Telescope at the Xuyi Observation Station. The primary goal of this collaboration was to discover super-fast rotators (SFRs) and study their physical properties. Two campaigns have been conducted: (a) a survey of ∼45 deg{sup 2} using 8 minute cadence during 2017 February 26–March 2, and (b) a survey of ∼60 deg{sup 2} using 10 minute cadence during 2018 March 9–12. Our samples are mainly main-belt asteroids and some Hildas and Jupiter Trojans. Out of 4522 collected lightcurves, 506 reliable rotation periods were obtained. Among the reliable rotation periods, we found 16 candidates with a possible rotation period of <2.2 hr, in which (134291) 2006 DZ6 shows a very convincing folded lightcurve and the other 15 candidates only have a likely trend. Further confirmation is needed for the rotation periods of these SFR candidates. In addition, (2280) Kunikov seems to have an eclipsing feature on its lightcurve with a relatively long rotation period suggesting that it is likely a fully synchronized binary asteroid. When the preliminary spin-rate distributions were carried out for asteroids using different sizes, no obvious difference was found.

79 ASTRONOMY AND ASTROPHYSICS↗

In-situ determination of strain during transient burst testing and the temperature dependence of Zircaloy-4 claddings

Understanding fuel system behavior during postulated loss-of-coolant accidents is pertinent for continued safe and efficient operation of light water reactors, particularly as higher burnups are being pursued and safety margins re-evaluated. Conventional mechanical models for the incumbent Zr alloys typically rely on the assumption that steady-state creep is the dominant fuel cladding response during transient accident conditions. To investigate this assumption, simulated accident burst testing was performed on Zircaloy-4 claddings with balloon behavior measured in-situ. Here, two distinct loading conditions were utilized during burst testing: (1) constant-gas-inventory where pressure was allowed to increase with temperature and (2) constant pressure. In-situ strains and strain rates were measured via 2-dimensional digital image correlation techniques and synchronized with temperature to determine deformation dependencies. The temperature dependence of strain rate was characterized by a two segment Arrhenius relationship, with a distinct transition between the high and low temperature/strain regimes. The average activation energy of the lower temperature/strain regime was 328 ± 25 kJ/mol, in agreement with the ~320 kJ/mol used for conventional LOCA models. However, the higher temperature/strain segment, which encompassed most of ballooning, showed increased activation energies as well as a dependence on whether the burst region was in view. For tests that burst away from the camera view, the average high temperature/strain segment activation energy was 635 ± 150 kJ/mol. For samples where the rupture opening formed in view, the average activation energy was 1015 ± 179 kJ/mol. This observed shift in temperature dependence indicates a transition in deformation mechanism at the end of life, possibly to time independent failure mechanisms, which has not yet been visualized in the literature for Zr alloys. Parameters at the transition points were analyzed to determine thresholds for this change in behavior, which occurred at an average hoop strain of 6.9 ± 2.1 %.

36 MATERIALS SCIENCE↗

MO-FG-202-03: Efficient Data Collection of Continuous 2D and Discrete Relative Dosimetric Data for Annual LINAC QA Using TrueBeam Developer Mode and a 1D Scanning Tank

Purpose: To develop a method to exploit real-time dynamic machine and couch parameter control during linear accelerator (LINAC) beam delivery to facilitate efficient performance of TG-142 suggested, Annual LINAC QA tests. Methods: Varian’s TrueBeam Developer Mode (Varian Medical Systems, Palo Alto, CA) facilitates control of Varian’s TrueBeam LINAC via instructions provided in Extensible Markup Language (XML) files. This allows machine and couch parameters to be varied dynamically, in real-time, during beam delivery. Custom XML files were created to allow for the collection of (1) continuous Tissue Maximum Ratios (TMRs), (2) beam profiles, and (3) continuous output factors using a 1D-scanning tank. TMRs were acquired by orienting an ionization chamber (IC) at isocenter (depth=25cm) and synchronizing a depth scan towards the water surface while lowering the couch at 1mm/s. For beam profiles, the couch was driven laterally and longitudinally while logging IC electrometer readings. Output factors (OFs) where collected by continually varying field sizes (4×4 to 30×30-cm{sup 2}) at a constant speed of 6.66 mm/s. To validate measurements, comparisons were made to data collected using traditional methods (e.g. 1D or 3D tank). Results: All data collecting using the proposed methods agreed with traditionally collected data (TMRs within 1%, OFs within 0.5% and beam profile agreement within 1% / 1mm) while taking less time to collect (factor of approximately 1/10) and with a finer sample resolution. Conclusion: TrueBeam developer mode facilitates collection of continuous data with the same accuracy as traditionally collected data with a finer resolution in less time. Results demonstrate an order of magnitude increase in sampled resolution and an order of magnitude reduction in collection time compared to traditional acquisition methods (e.g. 3D scanning tank). We are currently extending this approach to perform other TG-142 tasks.

61 RADIATION PROTECTION AND DOSIMETRY↗

Computer Science Research Needs for Parallel Discrete Event Simulation (PDES)

Historically, scientific computing efforts have demonstrated the clear need for, and effective use of, supercomputing with traditional time-stepped simulations. Nevertheless, there are several areas in the mission spaces of the U.S. Department of Energy and other agencies waiting to tap advanced computing research using a different, discrete event style of modeling, simulation, and analysis. These span a wide spectrum of applications including energy grid resilience, urban planning and policy, transportation science, building technologies, emergency response and planning, environmental impact analysis, computational epidemiology, Internet communications, cyber security, and cyber-physical systems, to name only a few. Even within traditional scientific applications, the role of discrete event modes of execution is increasing in the form of new event-based mathematical solvers such as quantized state integration methods and discrete-continuous hybrid system solvers. Co-design of advanced supercomputing hardware systems is another area that exploits discrete event simulation at its core for effective analyses. Complex systems, entity behaviors and interconnections play a significant role in all these applications, which are mapped to large-scale models with discrete event formulations. To make advancements in all the aforementioned scientific areas, many technical aspects need to be more thoroughly studied and deeply understood in parallel discrete event simulation (PDES). The unique dynamics inherent in a discrete event modeling approach, by their very nature, intersect and influence the entire stack of the computing system, including (a) the unique nature of the instruction sets exercised in PDES workloads without a predominance of high-precision floating point operations, (b) virtual time-constrained multi-threaded execution of many logical processes per processor, (c) extremely variable and difficult to predict network traffic characteristics, (d) interfaces and inter-dependencies with machine learning and artificial intelligence codes at higher software layers, and (e) highly challenging load balancing needs, especially in effectively accounting for accelerated/extremely heterogeneous computing in current and future high-performance computing systems. Efficient and accurate parallel execution of PDES workloads is also dominated by challenges in dealing with their asynchronous concurrency fundamentally present at the model level. Conservative synchronization, optimistic/speculative synchronization, and their hybrid schemes open new questions in fundamental computer science with respect to reversibility of computation and prediction (lookahead) of behaviors inherent within model codes. On the implementation front, there are relatively few scalable, general-purpose parallel discrete event simulators in the world, and even fewer have been studied on emerging hardware platforms. To enable scientific advances using PDES, the research needs in computer science must also be pursued and met in the intersection of the algorithmic and hardware-aware aspects of scalable PDES engines. This report is aimed at capturing a computer science-oriented view of this important area of research in PDES, presenting a sample of important applications with their inherent discrete event technology elements. Needs are outlined in core areas of parallel discrete event research as well as cross-cutting directions in computer science research that positively impact scientific advancements across several important application areas. A selection of priority research opportunities in advanced computing for PDES is identified to serve as reference for key research topics and their order of importance for scientific advancements.

97 MATHEMATICS AND COMPUTING↗

Distributed-Memory Sparse Deep Neural Network Inference Using Global Arrays

Partitioned Global Address Space (PGAS) models exhibit tremendous promise in developing efficient and productive distributed-memory parallel applications. They have been used extensively in scientific computations due to conveniently offering a ``shared-memory''-like model and convenient interfaces that separate communication with synchronization. Traditionally, PGAS communication models have been applied to dense/contiguously distributed data, but most modern applications depict varied levels of sparsity. Existing PGAS models require certain adaptations to support distributed sparse computations, since associated computations often require matrix arithmetic, in addition to data movement. The Global Arrays toolkit from Pacific Northwest National Laboratory (PNNL) is one of the earliest PGAS models to combine one-sided data communication and distributed matrix operations and is still used in the popular NWChem quantum chemistry suite. Recently, we have expanded the Global Arrays toolkit to support common sparse operations, like sparse matrix-dense matrix multiplies (SpMM), sparse matrix-sparse matrix multiplication (SpGEMM) and Sampled Dense-Dense Matrix Multiplication (SDDMM). As it turns out, these operations are the bedrock of sparse Deep Learning (DL); sparse deep neural networks and Graph Neural Networks (GNNs) have gained increasing attention recently in achieving speedups on training and inference with reduced memory footprints. Unlike scientific applications in High Performance Computing (HPC), modern (distributed-memory capable) DL toolkits often rely on non-standardized and closed-source vendor software optimizations, creating challenges in software-hardware co-design at scale. Our goal is to support a variety of distributed-memory sparse matrix operations and helper functions in the newly created Sparse Global Arrays (SGA), such that it is possible to build portable and productive Machine Learning scenarios for algorithm/software and hardware codesign purposes. Contemporary data-parallel schemes for training/inference are undergoing a major overhaul since model replication limits scalability and causes resource inefficiencies. As such, we have adopted tensor parallelism in decomposing the model and inputs, to mitigate memory issues. Current implementation is built on top of MPI and uses CPUs to maximize the portability across the platforms.

Distributed computing, machine learning↗

Application of high-energy x-rays and pair-distribution-function analysis to nano-scale structural studies in catalysis.

We investigate the structure of supported Pt catalysts using high-energy X-ray scattering coupled with Pair-Distribution-Function (PDF) analysis. Recently, experimental approaches that enable the collection of PDF data in situ have been developed with time-resolution sufficient to study the structure of Pt nano-particles as they form. The differential PDF approach is utilized which allows the atom-atom correlations involving only Pt to be selectively recovered, enabling structural investigation of the supported particles and the mechanism of their formation. In parallel to the in situ analysis, we have examined samples prepared ex situ. Data collected on the ex situ samples show that the initial deposition of Pt{sup 4+} occurs as the PtCl{sub 6}{sup 2-} species which are retained even when annealed in an oxygen atmosphere. The Pt differential PDFs of the samples reduced in hydrogen at 200 and 500 C indicated nano-crystalline face-centered-cubic (fcc) metallic Pt particles. The ex situ reduced samples also contain a weak correlations at 2.1 {angstrom}, which we assign to Pt-O interactions between the particles and the support surface. The in situ experiments, following the reduction of Pt{sup 4+} from 0 to 227 C, indicate that the initial Pt nano-particles formed are ca. 1 nm in size, and become larger and more crystalline by 200 C. The data suggest a particle growth mechanism where the initial particles that form are small (<1 nm), then agglomerate into ensembles of many small particles and lastly anneal to form larger well-ordered particles. Lastly, we discus potential future developments in operando PDF studies, and identify opportunities for synchronous application of complementary methods.

08 HYDROGEN↗

Photon diagnostics at the FLASH THz beamline

The THz beamline at FLASH, DESY, provides both tunable (1–300 THz) narrow-bandwidth (&sim;10%) and broad-bandwidth intense (up to 150 uJ) THz pulses delivered in 1 MHz bursts and naturally synchronized with free-electron laser X-ray pulses. Combination of these pulses, along with the auxiliary NIR and VIS ultrashort lasers, supports a plethora of dynamic investigations in physics, material science and biology. The unique features of the FLASH THz pulses and the accelerator source, however, bring along a set of challenges in the diagnostics of their key parameters: pulse energy, spectral, temporal and spatial profiles. Here, these challenges are discussed and the pulse diagnostic tools developed at FLASH are presented. In particular, a radiometric power measurement is presented that enables the derivation of the average pulse energy within a pulse burst across the spectral range, jitter-corrected electro-optical sampling for the full spectro-temporal pulse characterization, spatial beam profiling along the beam transport line and at the sample, and a lamellar grating based Fourier transform infrared spectrometer for the on-line assessment of the average THz pulse spectra. Corresponding measurement results provide a comprehensive insight into the THz beamline capabilities.

47 OTHER INSTRUMENTATION↗

Multiscale spectral nanoscopy

A system and method for non-invasively tracking a particle in a sample is disclosed. The system includes a 2-photon or confocal laser scanning microscope (LSM) and a particle-holding device coupled to a stage with X-Y and Z position control. The system also includes a tracking module having a tracking excitation laser, X-Y and Z radiation-gathering components configured to detect deviations of the particle in an X-Y and Z directions. The system also includes a processor coupled to the X-Y and Z radiation gathering components, generate control signals configured to drive the stage X-Y and Z position controls to track the movement of the particle. The system may also include a synchronization module configured to generate LSM pixels stamped with stage position and a processing module configured to generate a 3D image showing the 3D trajectory of a particle using the LSM pixels stamped with stage position.

42 ENGINEERING↗

Multi-channel lock-in amplifier assisted femtosecond time-resolved fluorescence non-collinear optical parametric amplification spectroscopy with efficient rejection of superfluorescence background

Superfluorescence appears as an intense background in femtosecond time-resolved fluorescence noncollinear optical parametric amplification spectroscopy, which severely interferes the reliable acquisition of the time-resolved fluorescence spectra especially for an optically dilute sample. Superfluorescence originates from the optical amplification of the vacuum quantum noise, which would be inevitably concomitant with the amplified fluorescence photons during the optical parametric amplification process. Here, we report the development of a femtosecond time-resolved fluorescence non-collinear optical parametric amplification spectrometer assisted with a 32-channel lock-in amplifier for efficient rejection of the superfluorescence background. With this spectrometer, the superfluorescence background signal can be significantly reduced to 1/300–1/100 when the seeding fluorescence is modulated. An integrated 32-bundle optical fiber is used as a linear array light receiver connected to 32 photodiodes in one-to-one mode, and the photodiodes are further coupled to a home-built 32-channel synchronous digital lock-in amplifier. As an implementation, time-resolved fluorescence spectra for rhodamine 6G dye in ethanol solution at an optically dilute concentration of 10{sup −5}M excited at 510 nm with an excitation intensity of 70 nJ/pulse have been successfully recorded, and the detection limit at a pump intensity of 60 μJ/pulse was determined as about 13 photons/pulse. Concentration dependent redshift starting at 30 ps after the excitation in time-resolved fluorescence spectra of this dye has also been observed, which can be attributed to the formation of the excimer at a higher concentration, while the blueshift in the earlier time within 10 ps is attributed to the solvation process.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Using scalable computer vision to automate high-throughput semiconductor characterization

Abstract High-throughput materials synthesis methods, crucial for discovering novel functional materials, face a bottleneck in property characterization. These high-throughput synthesis tools produce 10 4 samples per hour using ink-based deposition while most characterization methods are either slow (conventional rates of 10 1 samples per hour) or rigid (e.g., designed for standard thin films), resulting in a bottleneck. To address this, we propose automated characterization (autocharacterization) tools that leverage adaptive computer vision for an 85x faster throughput compared to non-automated workflows. Our tools include a generalizable composition mapping tool and two scalable autocharacterization algorithms that: (1) autonomously compute the band gaps of 200 compositions in 6 minutes, and (2) autonomously compute the environmental stability of 200 compositions in 20 minutes, achieving 98.5% and 96.9% accuracy, respectively, when benchmarked against domain expert manual evaluation. These tools, demonstrated on the formamidinium (FA) and methylammonium (MA) mixed-cation perovskite system FA 1−x MA x PbI 3 , 0 ≤ x ≤ 1, significantly accelerate the characterization process, synchronizing it closer to the rate of high-throughput synthesis.

Science & Technology - Other Topics↗

Time differentiated nuclear resonance spectroscopy coupled with pulsed laser heating in diamond anvil cells

Developments in pulsed laser heating applied to nuclear resonance techniques are presented together with their applications to studies of geophysically relevant materials. Continuous laser heating in diamond anvil cells is a widely used method to generate extreme temperatures at static high pressure conditions in order to study the structure and properties of materials found in deep planetary interiors. The pulsed laser heating technique has advantages over continuous heating, including prevention of the spreading of heated sample and/or the pressure medium and, thus, a better stability of the heating process. Time differentiated data acquisition coupled with pulsed laser heating in diamond anvil cells was successfully tested at the Nuclear Resonance beamline (ID18) of the European Synchrotron Radiation Facility. We show examples applying the method to investigation of an assemblage containing ε-Fe, FeO, and Fe{sub 3}C using synchrotron Mössbauer source spectroscopy, FeCO{sub 3} using nuclear inelastic scattering, and Fe{sub 2}O{sub 3} using nuclear forward scattering. These examples demonstrate the applicability of pulsed laser heating in diamond anvil cells to spectroscopic techniques with long data acquisition times, because it enables stable pulsed heating with data collection at specific time intervals that are synchronized with laser pulses.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

On the spectral behaviour of the turbulence-driven power fluctuations of horizontal-axis turbines

In this article we consider the spectral behaviour of turbulence-driven power fluctuations for a single horizontal-axis turbine. To this end, a small-scale instrumented axial-flow hydrokinetic turbine model ( diameter=0.724 m ) is deployed in the long water flume situated in the laboratory facilities of IFREMER in Boulogne-sur-Mer, France, and synchronous measurements of the upstream velocity and the rotor are collected for different tip-speed ratios. The study confirms previous findings suggesting that the power spectra follow the velocity spectra behaviour in the large scales region and a steeper power law slope behaviour ( -11/3 ) over the inertial frequency sub-range. However, we show that both the amplitude of the power spectra and low-pass filtering effect over the inertial sub-range also depend on the rotor aero/hydrodynamics (e.g. dC L /dα) and the approaching flow deceleration and not solely on the rotational effects. In addition, we present a novel semi-analytical model to predict the dominant blade-passing frequency harmonics in the high-frequency regime using the rotationally sampled spectra technique. For all calculations, the distortion of incoming turbulence is taken into account.

42 ENGINEERING↗

SuperCDMS Event Reconstruction Using Convolutional Neural Networks [Thesis]

The SuperCDMS experiment uses cryogenic silicon and germanium detectors to search for dark matter candidates such as WIMPs (Weakly Interacting Massive Particles) streaming through the Earth. Collisions in the silicon and germanium crystals are expected to produce phonons whose thermal signatures can be measured. This thesis first describes the integration of a new Signal Distribution Unit (SDU) to the SuperCDMS data acquisition system, which allows for synchronization of multiple detectors and electronic/mechanical noise characterization via accelerometer, antenna, and AC phase measurements. From SuperCDMS detector data it is necessary to reconstruct the energies of the particle events. This thesis explores the use of Convolutional Neural Networks (CNNs) to perform this reconstruction and finds that, although they perform well, changing the noise model breaks the model and requires the neural network to be retrained. In order to mitigate this issue, a new CNN model is proposed which includes the noise Power Spectral Density (PSD) of the data as an additional input to the CNN. While it proves to be effective as a denoising algorithm, it still fails for data with a different noise model. However, including data from multiple PSDs in the neural network training sample allows it to handle data with different types of noise while still maintaining the quality of the reconstruction. Nevertheless, neural networks trained even on multiple PSDs do not robustly handle data taken with PSDs dissimilar to those in the training sample, suggesting that CNNs may need to be retrained whenever the noise environment changes in a significant way.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗