Search NASA⌕ Search

SEARCH · Search NASA

Results for “Input Sensitivity”

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 1,531 records · Page 85

Technical Data to Justify Full Burnup Credit in Criticality Safety Licensing Analyses

Enercon Services, Inc. (ENERCON) was requested under Task Order #2 to identify scientific and technical data needed to benchmark and justify Full Burnup Credit, which adds 16 fission products and 4 minor actinides to Actinide-Only burnup credit. The historical perspective for Full Burnup Credit is discussed, and interviews of organizations participating in burnup credit activities are summarized as a basis for identifying additional data needs and making recommendation. Input from burnup credit participants representing two segments of the commercial nuclear industry is provided. First, the Electric Power Research Institute (EPRI) has been very active in the development of Full Burnup Credit, representing the interests of nuclear utilities in achieving capacity gains for storage and transport casks. EPRI and its utility customers are interested in a swift resolution of the validation issues that are delaying the implementation of Full Burnup Credit [EPRI 2010b]. Second, used nuclear fuel storage and transportation Cask Vendors favor improving burnup credit beyond Actinide-Only burnup credit, although their discussion of specific burnup credit achievements and data needs was limited citing business sensitive and technical proprietary concerns. While Cask Vendor proprietary items are not specifically identified in this report, the needs of all nuclear industry participants are reflected in the conclusions and recommendations of this report. In addition, Oak Ridge National Laboratory (ORNL) and Sandia National Laboratory (SNL) were interviewed for their input into additional data needs to achieve Full Burnup Credit. ORNL was very open to discussions of Full Burnup Credit, with several telecoms and a visit by ENERCON to ORNL. For many years, ORNL has provided extensive support to the NRC regarding burnup credit in all of its forms. Discussions with ORNL focused on potential resolutions to the validation issues for the use of fission products. SNL was helpful in ENERCON's understanding of the difficult issues related to obtaining and analyzing additional cross section test data to support Full Burnup Credit. A PIRT (Phenomena Identification and Ranking Table) analysis was performed by ENERCON to evaluate the costs and benefits of acquiring different types of nuclear data in support of Full Burnup Credit. A PIRT exercise is a formal expert elicitation process with the final output being the ranking tables. The PIRT analysis (Table 7-4: Results of PIRT Evaluation) showed that the acquisition of additional Actinide-Only experimental data, although beneficial, was associated with high cost and is not necessarily needed. The conclusion was that the existing Radiochemical Assay (RCA) data plus the French Haut Taux de Combustion (HTC) and handbook Laboratory Critical Experiment (LCE) data provide adequate benchmark validation for Actinide-Only Burnup Credit. The PIRT analysis indicated that the costs and schedule to obtain sufficient additional experimental data to support the addition of 16 fission products to Actinide-Only Burnup Credit to produce Full Burnup Credit are quite substantial. ENERCON estimates the cost to be $50M to $100M with a schedule of five or more years. The PIRT analysis highlights another option for fission product burnup credit, which is the application of computer-based uncertainty analyses (S/U - Sensitivity/Uncertainty methodologies), confirmed by the limited experimental data that is already available. S/U analyses essentially transform cross section uncertainty information contained in the cross section libraries into a reactivity bias and uncertainty. Recent work by ORNL and EPRI has shown that a methodology to support Full Burnup Credit is possible using a combination of traditional RCA and LCE validation plus S/U validation for fission product isotopics and cross sections. Further, the most recent cross section data (ENDF/B-VII) can be incorporated into the burnup credit codes at a reasonable cost compared to the acquisition of equivalent experimental data. ENERCON concludes that even with the costs of code data library updating, the use of S/U analysis methodologies could be accomplished on a shorter schedule and a lower cost than the gathering of sufficient experimental data. ENERCON estimates of the costs of an updated S/U computer code and data suite are $5M to $10M with a schedule of two to three years. Recent ORNL analyses using the S/U analysis method show that the bias and uncertainty values for fission product cross sections are smaller than previously expected. This result is confirmed by a similar EPRI approach using different data and computer codes. ENERCON also found that some issues regarding the implementation of burnup credit appear to have been successfully resolved especially the axial burnup profile issue and the depletion parameter issue. These issues were resolved through data gathering activities at the Yucca Mountain Project and ORNL.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

The challenges of detecting and attributing ocean acidification impacts on marine ecosystems

Abstract A substantial body of research now exists demonstrating sensitivities of marine organisms to ocean acidification (OA) in laboratory settings. However, corresponding in situ observations of marine species or ecosystem changes that can be unequivocally attributed to anthropogenic OA are limited. Challenges remain in detecting and attributing OA effects in nature, in part because multiple environmental changes are co-occurring with OA, all of which have the potential to influence marine ecosystem responses. Furthermore, the change in ocean pH since the industrial revolution is small relative to the natural variability within many systems, making it difficult to detect, and in some cases, has yet to cross physiological thresholds. The small number of studies that clearly document OA impacts in nature cannot be interpreted as a lack of larger-scale attributable impacts at the present time or in the future but highlights the need for innovative research approaches and analyses. We summarize the general findings in four relatively well-studied marine groups (seagrasses, pteropods, oysters, and coral reefs) and integrate overarching themes to highlight the challenges involved in detecting and attributing the effects of OA in natural environments. We then discuss four potential strategies to better evaluate and attribute OA impacts on species and ecosystems. First, we highlight the need for work quantifying the anthropogenic input of CO2 in coastal and open-ocean waters to understand how this increase in CO2 interacts with other physical and chemical factors to drive organismal conditions. Second, understanding OA-induced changes in population-level demography, potentially increased sensitivities in certain life stages, and how these effects scale to ecosystem-level processes (e.g. community metabolism) will improve our ability to attribute impacts to OA among co-varying parameters. Third, there is a great need to understand the potential modulation of OA impacts through the interplay of ecology and evolution (eco–evo dynamics). Lastly, further research efforts designed to detect, quantify, and project the effects of OA on marine organisms and ecosystems utilizing a comparative approach with long-term data sets will also provide critical information for informing the management of marine ecosystems.

Fisheries↗

Boundary layer loss sensitivity study using a modified ICRPG turbulent boundary layer computer program

Modifications of the turbulent boundary layer (TBL) computer program refer to a more accurate representation of boundary layer edge conditions, internal calculation of the Prandtl number, a changed friction coefficient relationship, and computation of the performance degradation. Important input parameters of the modified TBL program such as wall temperature distribution, Prandtl number, Stanton number, and velocity profile exponent were changed and the individual effects on significant boundary layer parameters, heat transfer, and performance degradation are described.

Omori, S.↗

Generic Sensor Modeling Using Pulse Method

Recent development of high spatial resolution satellites such as IKONOS, Quickbird and Orbview enable observation of the Earth's surface with sub-meter resolution. Compared to the 30 meter resolution of Landsat 5 TM, the amount of information in the output image was dramatically increased. In this era of high spatial resolution, the estimation of spatial quality of images is gaining attention. Historically, the Modulation Transfer Function (MTF) concept has been used to estimate an imaging system's spatial quality. Sometimes classified by target shapes, various methods were developed in laboratory environment utilizing sinusoidal inputs, periodic bar patterns and narrow slits. On-orbit sensor MTF estimation was performed on 30-meter GSD Landsat4 Thematic Mapper (TM) data from the bridge pulse target as a pulse input . Because of a high resolution sensor s small Ground Sampling Distance (GSD), reasonably sized man-made edge, pulse, and impulse targets can be deployed on a uniform grassy area with accurate control of ground targets using tarps and convex mirrors. All the previous work cited calculated MTF without testing the MTF estimator's performance. In previous report, a numerical generic sensor model had been developed to simulate and improve the performance of on-orbit MTF estimating techniques. Results from the previous sensor modeling report that have been incorporated into standard MTF estimation work include Fermi edge detection and the newly developed 4th order modified Savitzky-Golay (MSG) interpolation technique. Noise sensitivity had been studied by performing simulations on known noise sources and a sensor model. Extensive investigation was done to characterize multi-resolution ground noise. Finally, angle simulation was tested by using synthetic pulse targets with angles from 2 to 15 degrees, several brightness levels, and different noise levels from both ground targets and imaging system. As a continuing research activity using the developed sensor model, this report was dedicated to MTF estimation via pulse input method characterization using the Fermi edge detection and 4th order MSG interpolation method. The relationship between pulse width and MTF value at Nyquist was studied including error detection and correction schemes. Pulse target angle sensitivity was studied by using synthetic targets angled from 2 to 12 degrees. In this report, from the ground and system noise simulation, a minimum SNR value was suggested for a stable MTF value at Nyquist for the pulse method. Target width error detection and adjustment technique based on a smooth transition of MTF profile is presented, which is specifically applicable only to the pulse method with 3 pixel wide targets.

Helder, Dennis L.↗

Fully Distributed Multi-Material Magnetic Sensing Structures for Multiparameter DAS Applications

This dissertation demonstrates the first of its kind distributed magnetic field sensor based on a fiber optic distributed acoustic sensing (DAS) scheme. Ferromagnetic nickel and Metglas® were dispersed internally within a fiber optic preform and then drawn on an in-house fiber optic draw tower to lengths in the kilometers. Due to the close proximity of the ferromagnetic metals and fiber optic core, the magnetostrictive strain response of the ferromagnetic materials when exposed to a magnetic field would perturbate within the fiber cladding and transfer that strain, internally, to the fiber optic core. Strain resulting from the magnetostrictive effect allows the DAS based sensor to accurately translate strain into readable magnetic field data. Due to the high sensitivity seen in this sensor design, multiparameter sources, acoustic and magnetic fields, were tested and validated and a three dimensional magnetic-field vector sensor was proposed. Numerical analysis of the novel sensor design was first implemented using COMSOL Multiphysics, where inputs such as magnetostrictive element shape, size, distance, and number were first investigated. Upon optimizing system constraints, the sensor design was further modified such that single mode operation was consistent across multiple fiber draws while retaining high strain transfer from the ferromagnetic elements to the fiber optic core. Ferromagnetic material selection was evaluated as a function of the saturation magnetostriction constants and a total of 4 modules were used to fully characterize the complex physics involved in this sensor design. All fabrication and testing were performed in-house using a full scale 3-story fiber draw tower and custom environmental testing stations to imitate naturally occurring events such as magnetic or acoustic point sources. A unique stacking method was used to embed ferromagnetic nickel and Metglas® into a fiber optic preform which when combined with a custom fiber draw process resulted in consistent multi-material fibers drawn to lengths of 1-km. In-house testing facilities included different types of electromagnetic generators, in addition to a soil test bed, and an outdoor test bed which allowed 100 meters of fiber to be tested simultaneously. All tested sensors demonstrated high strain transfer capabilities on the order of 0.01-10 μϵ depending on the materials used, ferromagnetic rod number, and core to metal spacing. Due to the sensitivity of the system the difference between AC and DC was distinct, and directional magnetostriction was studied. Transverse and longitudinal magnetic wave propagation was controlled through a solenoid and rectangular Helmholtz coil, both built in-house. A three-dimensional magnetic field vector sensor was proposed due to the success of the magnetic field sensor, and a design was proposed and initially tested to validate direction as a function of field strength and distance. To summarize, this dissertation explores the first fully distributed magnetic field sensor using DAS based techniques and one of the first multi-material fiber draw processes which can produce consistent single mode fiber up to 1-km. Due to extensive FEA modeling, multiple iterations of the magnetic sensor were fully characterized and an equation describing the relationship between sensor design and strain transfer has been created and validated experimentally. Multi-parameter tests including acoustic and magnetic fields were implemented and an algorithm was developed to separate the mixed signals. Finally, a test was performed to demonstrate the feasibility of sensing magnetic fields directionally. Cumulative results demonstrate a high-quality sensor alternative to current designs which may surpass other magnetic sensors due to innate multi-parameter capabilities, in addition to the inexpensive production cost and extremely long operating lengths.

02 PETROLEUM↗

An EPGPT-based approach for uncertainty quantification

Generalized Perturbation Theory (GPT) has been widely used by many scientific disciplines to perform sensitivity analysis and uncertainty quantification. This manuscript employs recent developments in GPT theory, collectively referred to as Exact-to-Precision Generalized Perturbation Theory (EPGPT), to enable uncertainty quantification for computationally challenging models, e.g. nonlinear models associated with many input parameters and many output responses and with general non-Gaussian parameters distributions. The core difference between EPGPT and existing GPT is in the way the problem is formulated. GPT formulates an adjoint problem that is dependent on the response of interest. It tries to capture via the adjoint solution the relationship between the response of interest and the constraints on the state variations. EPGPT recasts the problem in terms of a smaller set of what is referred to as the 'active' responses which are solely dependent on the physics model and the boundary and initial conditions rather than on the responses of interest. The objective of this work is to apply an EPGPT methodology to propagate cross-sections variations in typical reactor design calculations. The goal is to illustrate its use and the associated impact for situations where the typical Gaussian assumption for parameters uncertainties is not valid and when nonlinear behavior must be considered. To allow this demonstration, exaggerated variations will be employed to stimulate nonlinear behavior in simple prototypical neutronics models. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

A Generalized Anisotropic Hardening Rule Based on the Mroz Multi-Yield-Surface Model for Pressure Insensitive and Sensitive Materials

In this paper, a generalized anisotropic hardening rule based on the Mroz multi-yield-surface model is derived. The evolution equation for the active yield surface is obtained by considering the continuous expansion of the active yield surface during the unloading/reloading process. The incremental constitutive relation based on the associated flow rule is then derived for a general yield function. As a special case, detailed incremental constitutive relations are derived for the Mises yield function. The closed-form solutions for one-dimensional stress-plastic strain curves are also derived and plotted for the Mises materials under cyclic loading conditions. The stress-plastic strain curves show closed hysteresis loops under uniaxial cyclic loading conditions and the Masing hypothesis is applicable. A user material subroutine based on the Mises yield function, the anisotropic hardening rule and the constitutive relations was then written and implemented into ABAQUS. Computations were conducted for a simple plane strain finite element model under uniaxial monotonic and cyclic loading conditions based on the anisotropic hardening rule and the isotropic and nonlinear kinematic hardening rules of ABAQUS. The results indicate that the plastic response of the material follows the intended input stress-strain data for the anisotropic hardening rule whereas the plastic response depends upon the input strain ranges of the stress-strain data for the nonlinear kinematic hardening rule.

cyclic hardening rule↗

Exact-to-precision generalized perturbation for neutron transport calculation

This manuscript extends the exact-to-precision generalized perturbation theory (E{sub P}GPT), introduced previously, to neutron transport calculation whereby previous developments focused on neutron diffusion calculation only. The E{sub P}GPT collectively denotes new developments in generalized perturbation theory (GPT) that place premium on computational efficiency and defendable accuracy in order to render GPT a standard analysis tool in routine design and safety reactor calculations. EPGPT constructs a surrogate model with quantifiable accuracy which can replace the original neutron transport model for subsequent engineering analysis, e.g. functionalization of the homogenized few-group cross sections in terms of various core conditions, sensitivity analysis and uncertainty quantification. This is achieved by reducing the effective dimensionality of the state variable (i.e. neutron angular flux) by projection onto an active subspace. Confining the state variations to the active subspace allows one to construct a small number of what is referred to as the 'active' responses which are solely dependent on the physics model rather than on the responses of interest, the number of input parameters, or the number of points in the state phase space. (authors)

97 MATHEMATICS AND COMPUTING↗

Optical Control of Adaptive Nanoscale Domain Networks

Adaptive networks can sense and adjust to dynamic environments to optimize their performance. Understanding their nanoscale responses to external stimuli is essential for applications in nanodevices and neuromorphic computing. However, it is challenging to image such responses on the nanoscale with crystallographic sensitivity. Here, the evolution of nanodomain networks in (PbTiO 3 ) n /(SrTiO 3 ) n superlattices (SLs) is directly visualized in real space as the system adapts to ultrafast repetitive optical excitations that emulate controlled neural inputs. The adaptive response allows the system to explore a wealth of metastable states that are previously inaccessible. Their reconfiguration and competition are quantitatively measured by scanning x-ray nanodiffraction as a function of the number of applied pulses, in which crystallographic characteristics are quantitatively assessed by assorted diffraction patterns using unsupervised machine-learning methods. The corresponding domain boundaries and their connectivity are drastically altered by light, holding promise for light-programable nanocircuits in analogy to neuroplasticity. Phase-field simulations elucidate that the reconfiguration of the domain networks is a result of the interplay between photocarriers and transient lattice temperature. The demonstrated optical control scheme and the uncovered nanoscopic insights open opportunities for the remote control of adaptive nanoscale domain networks.

36 MATERIALS SCIENCE↗

Effects of input gradient regularization on neural networks time-series forecasting of thermal power systems

This study proposes using neural networks, specifically gated recurrent unit (GRU), long-short-term memory (LSTM), and transformer networks, to improve control strategies in a 450 MW coal-fired power plant. However, neural networks face issues of becoming overly dependent on just a few variables to make predictions, which negatively impacts control decisions that rely on the model to determine the value of all manipulated variables. The paper introduces regularization techniques, including noise injection and input gradient regularization, during the training phase. Here, the work presents novel contributions in adapting neural networks to control industrial systems and applying regularization techniques from computer vision to industrial process control. Results demonstrate the effectiveness of input gradient regularization in reducing model dependence on subsets of variables, emphasizing the balance between fidelity and controllability. Further exploration is recommended, including the development of recurrent transformers, closed-loop control testing, and a sensitivity analysis on computer models to provide further insight.

20 FOSSIL-FUELED POWER PLANTS↗

Spatial impacts of technological innovations on the levelized cost of energy for offshore wind power plants in the United States

Recent studies predict significant decreases in the future levelized cost of energy (LCOE) of offshore wind energy, much of which is attributed to anticipated cost reductions from technological innovation. This study evaluates the spatial variability of LCOE caused by technology-induced decreases in a range of capital, operational, and financial cost categories. A spatial cost model of fixed-bottom and floating offshore wind plants is used to model the impact across thousands of potential United States sites. A specified change in an individual turbine subsystem cost produces a range of LCOE outcomes due to the varying geospatial characteristics of the considered sites and the nonlinear, interactive dependency on these input parameters; for example, a 10.8% improvement in net capacity factor can reduce LCOE by between 6% and 20% at different sites. This work expands upon the existing offshore wind literature, which typically evaluates cost sensitivities at a single site and does not consider the spatial variance in LCOE. The results suggest that the impact of technological innovations can be considerable and should be considered on a spatial as well as temporal basis when prioritizing technology innovation research or funding decisions to advance offshore wind technologies in the United States.

17 WIND ENERGY↗

Shock initiation of the HMX-based explosive PBX 9012: Experiments, uncertainty analysis, and unreacted equation-of-state

In this study, shock initiation experiments have been carried out on the polymer-bonded explosive PBX 9012 [nominally 90% octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazine (HMX), 10% vinylidene–hexafluoropropylene copolymer (Viton A) by weight] in order to provide calibration data for the explosive’s unreacted equation-of-state (EOS) and shock-to-detonation transition for reactive burn rate calibration. The input pressures covered the range of 1.86–4.43 GPa. This provided run-to-detonation depths ranging from > 22.3 mm for the lowest pressure shot to 4.91 mm at the highest pressure. The relative shock sensitivity of PBX 9012 is compared to other HMX-explosives in terms of the Pop-plot, showing that the studied explosive is more sensitive than other similar HMX-based counterparts (with notable exceptions). The uncertainty in the shock velocity determinations from the shock tracker measurements are also investigated, yielding new uncertainty measures in the generated Hugoniot data and run-to-detonation coordinates. Finally, the unreacted equation-of-state is determined using a linear U s – u p Mie–Grüneisen relation and the Davis reactants EOS analytical form, the latter being more suited for reactive burn modeling.

36 MATERIALS SCIENCE↗

A method for measuring coupling coefficients between cores and corrections to mode propagation constants in multicore fibres

A new experimental method is proposed for reconstructing supermode profiles of a multicore fibre (MCF), small corrections to the propagation constants of its cores and coupling coefficients between the cores without phase-sensitive measurements. The method is based on intensity measurements in the cores at the output of several MCF segments differing in length, using several configurations of controlled light launch into the cores at the fibre input (in the simplest case, by sequentially exciting each core), followed by numerical treatment using an iterative algorithm. The use of the method is demonstrated experimentally for a seven-core fibre. Using numerical simulation, we assess the stability of the method and the effect of the number of measurements on the accuracy of fibre parameter reconstruction. (paper)

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

All-optical control of molecular fluorescence

We present a quantum electrodynamical procedure to demonstrate the all-optical control of molecular fluorescence. The effect is achieved on passage of an off-resonant laser beam through an optically activated system; the presence of a surface is not required. Following the derivation and analysis of the all-optical control mechanism, calculations are given to quantify the significant modification of spontaneous fluorescent emission with input laser irradiance. Specific results are given for molecules whose electronic spectra are dominated by transitions between three electronic levels, and suitable laser experimental methods are proposed. It is also shown that the phenomenon is sensitive to the handedness of circularly polarized throughput, producing a conferred form of optical activity.

74 ATOMIC AND MOLECULAR PHYSICS↗

Mass Ejection by Strange Star Mergers and Observational Implications

We determine the Galactic production rate of strangelets as a canonical input to calculations of the measurable cosmic ray flux of strangelets by performing simulations of strange star mergers and combining the results with recent estimates of stellar binary populations. We find that the flux depends sensitively on the bag constant of the MIT bag model of QCD and disappears for high values of the bag constant and thus more compact strange stars. In the latter case, strange stars could coexist with ordinary neutron stars as they are not converted by the capture of cosmic ray strangelets. An unambiguous detection of an ordinary neutron star would then not rule out the strange matter hypothesis.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Suppressing Thermal Energy Drift In The LLNL Flash X-Ray Accelerator Using Linear Disk Resistor Stacks

This paper addresses thermal drift in sodium thiosulfate liquid resistors and their replacement with linear disk resistors from HVR Advanced Power Components. Sodium thiosulfate resistors in the FXR induction linear accelerator application have a temperature coefficient of {approx}1.8%/C. The FXR Marx banks send an 8kJ pulse through eight 524 cm{sup 3} liquid resistors at a repetition rate of up to 1 every 45 seconds. Every pulse increases the temperature of the solution by {approx}0.4 C which produces a 0.7% change in resistance. The typical cooling rate is {approx}0.4 C per minute which results in {approx}0.1% energy drop per pulse during continuous pulsed operations. A radiographic accelerator is extraordinarily sensitive to energy variations. Changes in beam energy produce movement in beam transport, changes in spot size, and large dose variations. If self-heating were the only problem, we could predict the increase in input voltage required to compensate for the energy loss. However, there are other variables that influence the temperature of the resistors such as focus magnet heating, changes in room temperature, changes in cooling water, where the cell is located, etc. Additionally not all of the resistors have equivalent cooling rates and as many as 32 resistors are driven from a single power source. The FXR accelerator group elected to replace the sodium thiosulfate resistors with HVR Linear Disk Resistors in a stack type configuration. With data limited for these resistors when used in oil and at low resistance values, a full characterization needed to be performed. High currents (up to 15kA), high voltages (up to 400kV), and Fast Rise times (<10ns) made a resistor choice difficult. Other solid resistors have been tried and had problems at the connection points and with the fact that the resistivity changed as they absorbed oil. The selected HVR resistors have the advantage of being manufactured with the oil impregnated in to them so this characteristic is minimized while still offering the desired low temperature coefficient of resistance compared to sodium thiosulfate. The characterization experiments and comparison with the sodium thiosulfate liquid resistors will be fully discussed and the final design described.

42 ENGINEERING↗

Regression Convolutional Neural Network for Energy Estimation in NOvA

Regression Convolutional Neural Network for Energy Estimation in NOvA" Abstract: "NOvA (NuMI Off-Axis $\nu_e$ Appearance) is a long baseline neutrino experiment designed to measure neutrino oscillations over a distance of 810 km. NOvA employs a near and far detector to observe $\nu_\mu$ disappearance and $\nu_e$ appearance of neutrinos produced by the NuMI beam at Fermilab. Energy reconstruction is critical for precise measurements of neutrino oscillation parameters and cross sections, which are functions of neutrino energy. Energy estimation remains difficult due to the complexity of detector response and final state particle kinematics. We present a regression-based convolutional neural network (CNN) method that reconstructs neutrino and lepton energies based on raw pixel inputs for NOvA. The trained model is able to reconstruct event energy for different interaction modes and complex final states containing leptons and hadrons. Studies of regression CNN networks show improved energy resolution and reduced sensitivity to calibration scale uncertainties relative to traditional kinematics-based energy reconstruction techniques. The results demonstrate the potential of the regression CNN method for neutrino physics analyses by improving on standard kinematics-based reconstruction.

Zhao, Larry [UC, Irvine (main)]↗

Simulation of Electron-Proton Scattering Events by a Feature-Augmented and Transformed Generative Adversarial Network (FAT-GAN)

We apply generative adversarial network (GAN) technology to build an event generator that simulates particle production in electron-proton scattering that is free of theoretical assumptions about underlying particle dynamics. The difficulty of efficiently training a GAN event simulator lies in learning the complicated patterns of the distributions of the particles physical properties. We develop a GAN that selects a set of transformed features from particle momenta that can be generated easily by the generator, and uses these to produce a set of augmented features that improve the sensitivity of the discriminator. The new Feature-Augmented and Transformed GAN (FAT-GAN) is able to faithfully reproduce the distribution of final state electron momenta in inclusive electron scattering, without the need for input derived from domain-based theoretical assumptions. The developed technology can play a significant role in boosting the science of existing and future accelerator facilities, such as the Electron-Ion Collider.

Alanazi, Yasir↗