Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “COLLECTIVE MODEL”

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 1,495 records · Page 83

Annual Report for Los Alamos National Laboratory Technical Area 54, Area G Disposal Facility – Fiscal Year 2011

As a condition to the Disposal Authorization Statement issued to Los Alamos National Laboratory (LANL or the Laboratory) on March 17, 2010, a comprehensive performance assessment and composite analysis maintenance program must be implemented for the Technical Area 54, Area G disposal facility. Annual determinations of the adequacy of the performance assessment and composite analysis are to be conducted under the maintenance program to ensure that the conclusions reached by those analyses continue to be valid. This report summarizes the results of the fiscal year 2011 annual review for Area G. Revision 4 of the Area G performance assessment and composite analysis was issued in 2008 and formally approved in 2009. These analyses are expected to provide reasonable estimates of the long-term performance of Area G and, hence, the disposal facility's ability to comply with Department of Energy (DOE) performance objectives. Annual disposal receipt reviews indicate that smaller volumes of waste will require disposal in the pits and shafts at Area G relative to what was projected for the performance assessment and composite analysis. The future inventories are projected to decrease modestly for the pits but increase substantially for the shafts due to an increase in the amount of tritium that is projected to require disposal. Overall, however, changes in the projected future inventories of waste are not expected to compromise the ability of Area G to satisfy DOE performance objectives. The Area G composite analysis addresses potential impacts from all waste disposed of at the facility, as well as other sources of radioactive material that may interact with releases from Area G. The level of knowledge about the other sources included in the composite analysis has not changed sufficiently to call into question the validity of that analysis. Ongoing environmental surveillance activities are conducted at, and in the vicinity of, Area G. However, the information generated by many of these activities cannot be used to evaluate the validity of the performance assessment and composite analysis models because the monitoring data collected are specific to operational releases or address receptors that are outside the domain of the performance assessment and composite analysis. In general, applicable monitoring data are supportive of some aspects of the performance assessment and composite analysis. Several research and development (R and D) efforts have been initiated under the performance assessment and composite analysis maintenance program. These investigations are designed to improve the current understanding of the disposal facility and site, thereby reducing the uncertainty associated with the projections of the long-term performance of Area G. The status and results of R and D activities that were undertaken in fiscal year 2011 are discussed in this report. Special analyses have been conducted to determine the feasibility of disposing of specific waste streams, to address proposed changes in disposal operations, and to consider the impacts of changes to the models used to conduct the performance assessment and composite analysis. These analyses are described and the results of the evaluations are summarized in this report. The Area G disposal facility consists of Material Disposal Area (MDA) G and the Zone 4 expansion area. To date, all disposal operations at Area G have been confined to MDA G. Material Disposal Area G is scheduled to undergo final closure in 2015; disposal of waste in the pits and shafts is scheduled to end in 2013. In anticipation of the closure of MDA G, plans are being made to ship the majority of the waste generated at LANL to off-site locations for disposal. It is not clear at this time if waste that will be disposed of at LANL will be placed in Zone 4 or if disposal operations will move to a new location at the Laboratory. Separately, efforts to optimize the final cover used in the closure of MDA G are underway; a final cover design different than that adopted for the performance assessment and composite analysis will likely emerge from that investigation. All of these changes will require re-examination of the assumptions upon which the performance assessment and composite analysis are based and, in all likelihood, revision of those analyses.

54 ENVIRONMENTAL SCIENCES↗

Outcomes of the 2013 GTO Workshop on Geothermal Code Comparison

Pacific Northwest National Laboratory (PNNL) is supporting the Department of Energy (DOE) Geothermal Technologies Office (GTO) in organizing and executing a model comparison activity. This project is directed at testing, diagnosing differences, and demonstrating modeling capabilities of a worldwide collection of numerical simulators for evaluating geothermal technologies. A key element of the project was the planning and implementation of a one-day project kickoff workshop, held February 14, 2013 in Palo Alto, CA. The primary goals of the workshop were to 1) introduce the project and its objectives to potential participating team members, and 2) develop an initial set of test problem descriptions for use in the execution stage. This report summarizes the outcomes of the workshop.

15 GEOTHERMAL ENERGY↗

Review of searches for vector-like quarks, vector-like leptons, and heavy neutral leptons in proton–proton collisions at $\sqrt{s} = 13$ TeV at the CMS experiment

The LHC has provided an unprecedented amount of proton–proton collision data, bringing forth exciting opportunities to address fundamental open questions in particle physics. These questions can potentially be answered by performing searches for very rare processes predicted by models that attempt to extend the standard model of particle physics. The data collected by the CMS experiment in 2015–2018 at a center-of-mass energy of 13 TeV can be used to test the standard model with high precision and potentially uncover evidence for new particles or interactions. An interesting possibility is the existence of new fermions with masses ranging from the MeV to the TeV scale. Such new particles appear in many possible extensions of the standard model and are well motivated theoretically. New fermions may explain the appearance of three generations of leptons and quarks, the mass hierarchy across these generations, and the nonzero neutrino masses. In this report, the results of searches targeting vectorlike quarks, vector-like leptons, and heavy neutral leptons at the CMS experiment are summarized. The complementarity of current searches for each type of new fermion is discussed, and combinations of several searches for vector-like quarks are presented. The discovery potential for some of these searches at the High-Luminosity LHC is also discussed.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Load Composition Model Workflow (BPA TIP-371 Deliverable 1A)

This project is funded under Bonneville Power Administration (BPA) Strategic Partnership Project (SPP) 17-005 between BPA and SLAC National Accelerator Laboratory. The project in a BPA Technology Improvement Project (TIP) that builds on and validates the Composite Load Model developed by the Western Electric Coordinating Council's (WECC) Load Modeling Task Force (LMTF). The composite load model is used by the WECC Modeling and Validation Work Group to study the stability and security of the western electricity interconnection. The work includes development of load composition data sets, collection of load disturbance data, and model development and validation. This work supports reliable and economic operation of the power system. This report was produced for Deliverable 1A of the BPA TIP-371 Project entitled \TIP 371: Advancing the Load Composition Model". The deliverable documents the proposed work ow for the Composite Load Model, which provides the basis for the instrumentation, data acquisition, analysis and data dissemination activities addressed by later phases of the project.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Development and Evaluation of Distributed Energy Resource Device Models: Electric Vehicles, Electric Water Heaters, and Commercial Refrigeration Systems

In order to validate distributed energy resource (DER) models operated with grid services in the GMLC 1.4.2 team’s February (GMLC 2019a) and July (GMLC 2019b) reports, a test and measurement program using actual DER devices was conducted by national laboratories for three devices: (1) electric vehicles (EVs), (2) water heaters, and (3) commercial refrigeration. Test procedures were developed and carried out to identify the mathematical models and their parameters that describe the operational function, characterize the physics, and obey transient response of the devices. This report focuses on the experimental results obtained to develop and verify simulations of three specific EV models. This section summarizes elements of the mathematical model and shows the necessity data collected from the EVs while charging and discharging (driving). In the course of the study, we found it necessary to make some modifications to the model to deal with observed transient behavior. Section 1.2 details the model assumptions and equations, Section 1.3 provides details of the EVs tested, Section 1.4 outlines the test procedures developed for the projects, and Section 1.5 presents the experimental results obtained for the three different EVs tested and how these results compare to the models running the same test profiles.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Review of searches for vector-like quarks, vector-like leptons, and heavy neutral leptons in proton-proton collisions at $\sqrt{s}$ = 13 TeV at the CMS experiment

The LHC has provided an unprecedented amount of proton-proton collision data, bringing forth exciting opportunities to address fundamental open questions in particle physics. These questions can potentially be answered by performing searches for very rare processes predicted by models that attempt to extend the standard model of particle physics. The data collected by the CMS experiment in 2015-2018 at a center-of-mass energy of 13 TeV can be used to test the standard model with high precision and potentially uncover evidence for new particles or interactions. An interesting possibility is the existence of new fermions with masses ranging from the MeV to the TeV scale. Such new particles appear in many possible extensions of the standard model and are well motivated theoretically. New fermions may explain the appearance of three generations of leptons and quarks, the mass hierarchy across the generations, and the nonzero neutrino masses. In this report, the results of searches targeting vector-like quarks, vector-like leptons, and heavy neutral leptons at the CMS experiment are summarized. The complementarity of current searches for each type of new fermion is discussed, and combinations of several searches for vector-like quarks are presented. The discovery potential for some of these searches at the High-Luminosity LHC is also discussed.

FOS: Physical sciences↗

GR 290 (ROMANO’S STAR). II. LIGHT HISTORY AND EVOLUTIONARY STATE

We have investigated the past light history of the luminous variable star GR 290 (M33/V532, Romano’s Star) in the M33 galaxy, and collected new spectrophotometric observations in order to analyze links between this object, the LBV category, and the Wolf–Rayet stars of the nitrogen sequence. We have built the historical light curve of GR 290 back to 1901, from old observations of the star found in several archival plates of M33. These old recordings together with published and new data on the star allowed us to infer that for at least half a century the star was in a low luminosity state, with B ≃ 18–19, most likely without brighter luminosity phases. After 1960, five large variability cycles of visual luminosity were recorded. The amplitude of the oscillations was seen increasing toward the 1992–1994 maximum, then decreasing during the last maxima. The recent light curve indicates that the photometric variations have been quite similar in all the bands and that the B – V color index has been constant within ±0.1{sup m} despite the 1.5{sup m} change of the visual luminosity. The spectrum of GR 290 at the large maximum of 1992–94 was equivalent to late-B-type, while, during 2002–2014, it varied between WN10h-11h near the visual maxima to WN8h-9h at the luminosity minima. We have detected, during this same period, a clear anti-correlation between the visual luminosity, the strength of the He ii 4686 Å emission line, the strength of the 4600–4700 Å lines’ blend, and the spectral type. From a model analysis of the spectra collected during the whole 2002–2014 period, we find that the Rosseland radius R {sub 2/3}, changed between the minimum and maximum luminosity phases by a factor of three while T {sub eff} varied between about 33,000 and 23,000 K. We confirm that the bolometric luminosity of the star has not been constant, but has increased by a factor of ∼1.5 between minimum and maximum luminosity, in phase with the apparent luminosity variations. Presently, GR 290 falls in the H–R diagram close to WN8h stars and is probably younger than them. In the light of current evolutionary models of very massive stars, we find that GR 290 has evolved from an ∼60 M {sub ☉} progenitor star and should have an age of about four million years. From its physical charcteristics, we argue that GR 290 has left the LBV stage and is presently moving from the LBV stage to a Wolf–Rayet stage of a late nitrogen spectral type.

79 ASTRONOMY AND ASTROPHYSICS↗

Analyzing the Largest Spectroscopic Data Set of Hydrogen-poor Super-luminous Supernovae

Super-luminous supernovae (SLSNe) are tremendously luminous explosions whose power sources and progenitors are highly debated. Broad-lined SNe Ic (SNe Ic-bl) are the only type of SNe that are connected with long-duration gamma-ray bursts (GRBs). Studying the spectral similarity and difference between the populations of hydrogen-poor SLSNe (SLSNe Ic) and of hydrogen-poor stripped-envelope core-collapse SNe, in particular SNe Ic and SNe Ic-bl, can provide crucial observations to test predictions of theories based on various power source models and progenitor models. In this paper, we collected all of the published optical spectra of 32 SLSNe Ic, 21 SNe Ic-bl, as well as 17 SNe Ic, quantified their spectral features, constructed average spectra, and compared them in a systematic way using new tools we have developed. We find that SLSNe Ic and SNe Ic-bl, including those connected with GRBs, have comparable widths for their spectral features and average absorption velocities at all phases. Thus, our findings strengthen the connection between SLSNe Ic and GRBs. In particular, SLSNe Ic have average Fe ii λ 5169 absorption velocities of −15,000 ± 2600 km s{sup −1} at 10 days after peak, which are higher than those of SNe Ic by ∼7000 km s{sup −1} on average. SLSNe Ic also have significantly broader Fe ii λ 5169 lines than SNe Ic. Moreover, we find that such high absorption and width velocities of SLSNe Ic may be hard to explain with the interaction model, and none of the 13 SLSNe Ic with measured absorption velocities spanning over 10 days has a convincing flat velocity evolution, which is inconsistent with the magnetar model in one dimension. Lastly, we compare SN 2011kl, the first SN connected with an ultra-long GRB, with the mean spectrum of SLSNe Ic and of SNe Ic-bl.

79 ASTRONOMY AND ASTROPHYSICS↗

Deep learning model to detect various synchrophasor data anomalies

High-density synchrophasors provide valuable information for power grid situational awareness, operation and control. Unfortunately, due to factors including communication instability and hardware failure, their data quality can be greatly deteriorated by anomalies. Since the anomalies can impact the performance of the synchrophasor applications, it is of paramount significance to propose a model to detect anomalies in synchrophasor. In this study, a convolutional neural network model is established to detect and classify the anomalies in the synchrophasor measurements. Additionally, four types of anomalies observed in actual synchrophasors including erroneous patterns, random spikes, missing points and high-frequency interferences are considered in this study. The proposed model is extensively evaluated via field-collected measurements from the synchrophasor network in Jiangsu grid, China. The superior performance of the proposed model indicates the great potential of using deep learning for the detection of abnormal synchrophasor measurements.

42 ENGINEERING↗

Centrality and transverse momentum dependence of D 0 -meson production at mid-rapidity in Au + Au collisions at s NN = 200 GeV

Here, we report a new measurement of D 0 -meson production at mid-rapidity (| y | < 1) in Au + Au collisions at $\sqrt{s}$$_ {NN}$ = 200 GeV utilizing the heavy flavor tracker, a high resolution silicon detector at the STAR experiment. Invariant yields of D 0 mesons with transverse momentum p T ≲ 9 GeV / c are reported in various centrality bins (0–10%, 10–20%, 20–40%, 40–60%, and 60–80%). Blast-wave thermal models are used to fit the D 0 -meson p T spectra to study D 0 hadron kinetic freeze-out properties. The average radial flow velocity extracted from the fit is considerably smaller than that of light hadrons (π, K, and p), but comparable to that of hadrons containing multiple strange quarks (Φ, Ξ - ) , indicating that D 0 mesons kinetically decouple from the system earlier than light hadrons. The calculated D 0 nuclear modification factors reaffirm that charm quarks suffer a large amount of energy loss in the medium, similar to those of light quarks for p T > 4 GeV/c in central 0–10% Au + Au collisions. At low p T , the nuclear modification factors show a characteristic structure qualitatively consistent with the expectation from model predictions that charm quarks gain sizable collective motion during the medium evolution. The improved measurements are expected to offer new constraints to model calculations and help gain further insights into the hot and dense medium created in these collisions.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

OPERA models for predicting physicochemical properties and environmental fate endpoints

The collection of chemical structure information and associated experimental data for quantitative structure–activity/ property relationship (QSAR/QSPR) modeling is facilitated by an increasing number of public databases containing large amounts of useful data. However, the performance of QSAR models highly depends on the quality of the data and modeling methodology used. This study aims to develop robust QSAR/QSPR models for chemical properties of environmental interest that can be used for regulatory purposes. This study primarily uses data from the publicly available PHYSPROP database consisting of a set of 13 common physicochemical and environmental fate properties. These datasets have undergone extensive curation using an automated workflow to select only high-quality data, and the chemical structures were standardized prior to calculation of the molecular descriptors. The modeling procedure was developed based on the five Organization for Economic Cooperation and Development (OECD) principles for QSAR models. A weighted k-nearest neighbor approach was adopted using a minimum number of required descriptors calculated using PaDEL, an open-source software. The genetic algorithms selected only the most pertinent and mechanistically interpretable descriptors (2–15, with an average of 11 descriptors). The sizes of the modeled datasets varied from 150 chemicals for biodegradability half-life to 14,050 chemicals for logP, with an average of 3222 chemicals across all endpoints. The optimal models were built on randomly selected training sets (75%) and validated using fivefold cross-validation (CV) and test sets (25%). The CV Q2 of the models varied from 0.72 to 0.95, with an average of 0.86 and an R2 test value from 0.71 to 0.96, with an average of 0.82. Modeling and performance details are described in QSAR model reporting format and were validated by the European Commission’s Joint Research Center to be OECD compliant. All models are freely available as an open-source, command-line application called OPEn structure–activity/property Relationship App (OPERA). OPERA models were applied to more than 750,000 chemicals to produce freely available predicted data on the U.S. Environmental Protection Agency’s CompTox Chemistry Dashboard.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Modeling and Validation of a Residential Multi-Functional Variable Refrigerant Flow Heat Pump System with Heat Recovery

To bridge the existing gap in modeling the variable refrigerant flow heat pump systems with heat recovery (VRFHR), we developed a suite of dynamic VRFHR system models in Modelica. These models are specifically tailored for residential multi-functional VRFHR (MF-VRFHR) applications, including space conditioning and domestic hot water (DHW) heating, utilizing both the TIL library for HVAC equipment and the Buildings library for thermal load calculations. The development comprises essential component models, including the newly developed heat recovery unit (HRU), along with system models that integrate the heat pump system and building envelope. These system models accommodate various operational modes such as heating-only, cooling-only, and heating-recovery (including heating-dominant and cooling-dominant) modes. Furthermore, we propose an efficient optimization-based model calibration method that identifies critical model parameters while utilizing a small amount of data obtained from either real systems or manufacturer's specifications. We demonstrate the effectiveness of these models and the proposed calibration method for a MF-VRFHR system installed in Richland, WA. The developed models are calibrated and validated using data collected under different operational modes during both heating and cooling seasons. The results show that the models capture the system dynamics and achieve high accuracy, with the coefficient of the variation of the root-mean-square-error less than 15% for variables such as outdoor unit power consumption, compressor speed, space temperature and DHW temperature. The validated models serve as a reliable representation of the MF-VRFHR system, facilitating the development and validation of optimized controls needed to realize the full benefits of integrated heat pump systems. Future research will utilize these models to develop advanced controls and optimize system performance for improved energy efficiency and demand flexibility.

Modeling, Variable refrigerant flow (VRF) systems,↗

NASA's Electric Sail Propulsion System Investigations over the Past Three Years

Personnel from NASA's MSFC have been investigating the feasibility of an advanced propulsion system known as the Electric Sail for future scientific missions of exploration. This team initially won a NASA Space Technology Mission Directorate (STMD) Phase I NASA Innovative Advanced Concept (NIAC) award and then a two year follow-on Phase II NIAC award. This paper documents the findings from this three year investigation. An Electric sail propulsion system is a propellant-less and extremely fast propulsion system that takes advantage of the ions that are present in the solar wind to provide very rapid transit speeds whether to deep space or to the inner solar system. Scientific spacecraft could arrive to Pluto in ~5 years, to the boundary of the solar system in ten to twelve years vs. thirty five plus years it took the Voyager spacecraft. The team's recent focused activities are: 1) Developing a Particle in Cell (PIC) numeric engineering model from the experimental data collected at MSFC's Solar Wind Facility on the interaction between simulated solar wind interaction with a charged bare wire that can be applied to a variety of missions, 2) The development of the necessary tether deployers/tethers to enable successful deployment of multiple, multi km length bare tethers, 3) Determining the different missions that can be captured from this revolutionary propulsion system 4) Conceptual designs of spacecraft to reach various destinations whether to the edge of the solar system, or as Heliophysics sentinels around the sun, or to trips to examine a multitude of asteroids These above activities, once demonstrated analytically, will require a technology demonstration mission (~2021 to 2023) to demonstrate that all systems work together seamlessly before a Heliophysics Electrostatic Rapid Transit System (HERTS) could be given the go-ahead. The proposed demonstration mission will require that a small spacecraft must first travel to cis-lunar space as the Electric Sail must be outside of Earth's Magnetic fields to produce thrust. The paper will outline what was done over the past three years from performing various plasma chamber tests to obtain data for the PIC model development, investigation of tether material trades, and conceptual designs of proposed spacecraft.

Wiegmann, Bruce M.↗

Nucleon-pair approximation for nuclei from spherical to deformed regions

Here we model low-lying states of atomic nuclei in the nucleon-pair approximation of the shell model, using three approaches to select collective nucleon pairs: the generalized seniority scheme, the conjugate gradient method, and the Hartree-Fock approach. We find the collective pairs obtained from the generalized seniority scheme provide a good description for nearly spherical nuclei, and those from the conjugate gradient method or the Hartree-Fock approach work well for transitional and deformed nuclei. Our NPA calculations using collective pairs with angular momenta 0, 2, and 4 (denoted by S D G pairs) reproduce the nuclear shape evolution in the N = 26 isotones, Ca 46 , Ti 48 , Cr 50 , and Fe 52 , and yield good agreement with full configuration-interaction calculations of low-lying states in medium-heavy transitional and deformed nuclei: Ti 44 – 48 , Cr 48 , Cr 50 , Fe 52 , Zn 60 – 64 , Ge 64 , 66 , Mo 84 , and Xe 108 – 112 . Finally, using the S D G I -pair approximation we describe low-lying states of Ba 112 , 114 , cases difficult to reach by conventional configuration-interaction methods.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Nutrient Cycle Impacts on Forest Ecosystem Carbon Cycling: Improved Prediction of Climate Feedbacks from Coupled C–Nutrient Dynamics from Ecosystem to Regional Scales

One of the largest sources of uncertainty in the projected global mean temperature response to anthropogenic carbon (C) emissions is the role of the land surface in contributing to or mediating the rise in atmospheric CO2 [IPCC, 2007]. Development of ecosystem processes in global land surface models (LSMs) is far from complete, with major components such as the nitrogen (N) cycle only recently being implemented in some LSMs [Thornton et al., 2007; Sokolov et al., 2008; Xu-Ri and Prentice, 2008; Fisher et al., 2010]. Inclusion of the N cycle can reduce terrestrial C uptake by up to ~50 Pg C·y-1 globally [Figure 1: Fisher et al., 2010]. However, even these early representations of the N cycle in LSMs are missing critical plant–soil interactions that have been shown empirically to exert a major influence over ecosystem C fluxes. Here we propose to use a newly developed plant N model by PI Fisher as a theoretical and analytical framework for integrating novel field-based methods and long-term flux observations to quantify the impacts of belowground processes on fluxes of C, water and energy, from ecosystem to regional scales. Although the model (Fixation & Uptake of Nitrogen: FUN) is one of the more advanced global plant N models available, the model is dependent on theoretical C flux curves optimizing plant C allocation to root exudates, symbiotic bacteria and fungi, and retranslocation enzymes that have not been confirmed or validated with measurements—primarily because the measurement techniques had not been available, until recently. Parallel to the model developments by Fisher et al. were developments in in situ direct measurement techniques by Phillips et al. [2008; 2011], who were able to measure C fluxes from roots to soil. Not only do these techniques have the capacity to confirm and/or modify the modeling components in the Fisher et al. plant N model, but they also provide insight into how belowground C supply from roots couples the C–N cycles at the micro-site and ecosystem scales. Collectively, the integration of measurements and modeling provides empirical grounding for theoretical process behavior. Three critical components will enable us to successfully carry out this research: I) new methodological approaches developed by co-I Phillips for in situ measurement of root exudates, and belowground C fluxes that allow for testing and refinement of FUN [Phillips et al., 2008; Phillips et al., 2011]; II) an innovative research design that involves data integration across an ecological gradient, including a DOE-supported AmeriFlux tower operated by co-I Dragoni, which is essential to determining the ecosystem flux impacts [Dragoni et al., 2011]; and, III) new results from co-I Evans showing unique spectral signatures of trees in association with ectomycorrhizae (ECM) and arbuscular mycorrhizae (AM) fungi from Landsat data using spectral mixture analysis (SMA), which is key to scaling up belowground processes to regional scales [Andersson et al., 2009]. Each component provides feedback to and is intimately linked with all other components; no part operates in isolation. Our team and approach combines expertise in modeling, remote sensing, ecosystem ecology, plant physiology, soil biogeochemistry, and micrometeorology to address these inherently interdisciplinary problems and questions. This proposal directly responds to the focused Science Area of the FOA: “The role of belowground processes and mechanisms across scales (e.g., soil carbon transformation, root dynamics, mycorrhizal interactions, and plant mediated (e.g., root exudates) biogeochemical transformations) associated with a changing climate.

54 ENVIRONMENTAL SCIENCES↗

Analyzing inference workloads for spatiotemporal modeling

Ensuring power grid resiliency, forecasting climate conditions, and optimization of transportation infrastructure are some of the many application areas where data is collected in both space and time. Spatiotemporal modeling is about modeling those patterns for forecasting future trends and carrying out critical decision-making by leveraging machine learning/deep learning. Once trained offline, field deployment of trained models for near real-time inference could be challenging because performance can vary significantly depending on the environment, available compute resources and tolerance to ambiguity in results. Users deploying spatiotemporal models for solving complex problems can benefit from analytical studies considering a plethora of system adaptations to understand the associated performance-quality trade-offs. To facilitate the co-design of next-generation hardware architectures for field deployment of trained models, it is critical to characterize the workloads of these deep learning (DL) applications during inference and assess their computational patterns at different levels of the execution stack. In this paper, we develop several variants of deep learning applications that use spatiotemporal data from dynamical systems. We study the associated computational patterns for inference workloads at different levels, considering relevant models (Long short-term Memory, Convolutional Neural Network and Spatio-Temporal Graph Convolution Network), DL frameworks (Tensorflow and PyTorch), precision (FP16, FP32, AMP, INT16 and INT8), inference runtime (ONNX and AI Template), post-training quantization (TensorRT) and platforms (Nvidia DGX A100 and Sambanova SN10 RDU). Overall, our findings indicate that although there is potential in mixed-precision models and post-training quantization for spatiotemporal modeling, extracting efficiency from contemporary GPU systems might be challenging. Instead, co-designing custom accelerators by leveraging optimized High Level Synthesis frameworks (such as SODA High-Level Synthesizer for customized FPGA/ASIC targets) can make workload-specific adjustments to enhance the efficiency.

97 MATHEMATICS AND COMPUTING↗

Observation of Coherent 𝜙⁡(1020) Meson Photoproduction in Ultraperipheral PbPb Collisions at $\sqrt{𝑠_{NN}}$ = 5.36 TeV

The first observation of coherent 𝜙⁡(1020) meson photoproduction off heavy nuclei is presented using ultraperipheral lead-lead collisions at a center-of-mass energy per nucleon pair of 5.36 TeV. The data were collected by the CMS experiment and correspond to an integrated luminosity of 1.62 μ⁢𝑏 −1 . The 𝜙⁡(1020) meson signals are reconstructed via the 𝐾 + ⁢𝐾 − decay channel. The production cross section is presented as a function of the 𝜙⁡(1020) meson rapidity in the range 0.3 < |𝑦| < 1.0, probing gluons that carry a fraction of the nucleon momentum (𝑥) around 10 −4 . The observed cross section exhibits little dependence on rapidity and is significantly suppressed, by a factor of ∼5, compared to a baseline model that treats a nucleus as a collection of free nucleons. Theoretical models that incorporate the nuclear shadowing effect generally provide a better description of the 𝜙⁡(1020) data than those incorporating gluon saturation. This study establishes a powerful new tool for exploring nuclear effects and nuclear gluonic structure in the small-𝑥 regime at a unique energy scale bridging the perturbative and nonperturbative quantum chromodynamics domains.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗