Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Distributed Processing”

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 379 records · Page 21
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Measurements of Beam Spin Asymmetries in p+p0 and p´p0 Dihadron Production at CLAS12

Semi-Inclusive Deep Inelastic Scattering (SIDIS) is a powerful experimental tool for studying the internal structure and dynamics of the proton, revealing how quarks and gluons are distributed and interact within it. SIDIS describes a process where an elec tron scatters off one of the constituent quarks within the proton, causing it to undergo hadronization, creating multiple hadrons in the final state. Through factorization, the full process can be split into probabilistic components: one which describes the internal structure of the proton using Parton Distribution Functions (PDFs), and another which describes the hadronization process using Fragmentation Functions (FFs). These functions are non-perturbative quantities of Quantum Chromodynamics (QCD), meaning they cannot be calculated directly from first principles and must instead be extracted from experimental measurements. Acommon approach for accessing PDFs and FFs using SIDIS is to measure asymmetries. In this context, asymmetries correspond to subtle differences in the angular distribution of outgoing particles that arise when the spin orientation of the incoming beam or target is reversed. Because many of these effects only appear when spin is involved, they isolate specific, nuanced properties of the proton’s spin-structure that are otherwise hidden in spin averaged measurements. In practice, they show up as specific azimuthal modulations (e.g., sin ¿R, sin(¿h ´ ¿R)), whose amplitudes isolate convolutions of PDFs and FFs at leading and subleading twist. Non-zero asymmetries of these angular distributions can be traced back to unique combinations of PDFs and FFs, offering a way to probe them directly. In this work, we measure SIDIS by analyzing high energy electron-proton scattering events using the CLAS12 detector at Jefferson Lab. This study focuses on subset of SIDIS referred to as dihadron SIDIS, where pairs of hadrons — here p+p0 and p´p0 — are observed. We analyzed these dihadrons using detector data collected during Fall 2018 and Spring 2019, where longitudinally polarized electrons from the CEBAF accelerator were incident on a liquid hydrogen target. A photon classifier using a Gradient Boosted Trees (GBTs) architecture was trained using Monte Carlo simulations to reduce the amount of iv false combinatorial background p0’s. When deployed on experimental data, the model in creases our dihadron statistics by up to five-fold compared to previous CLAS12 p0 analyses. This work reports the first measurements of beam spin asymmetries for p+p0 and p´p0 dihadron production in SIDIS. The measured asymmetries offer new insights to the spin-dependent structure and dynamics within the proton, as well as the spin-dependent properties of quark fragmentation. Non-zero twist-3 sin¿R amplitudes are observed, pro viding sensitivity to the subleading twist PDF e(x). The PDF e(x) encodes quark-gluon correlations within the proton — a property that is otherwise inaccessible at leading twist. Additionally, this work measured significant twist-2 modulations carried by sin(¿h ´ ¿R) and sin(2¿h ´2¿R), providing experimental access to the helicity dihadron fragmentation function (DiFF) GK 1 . Because there is no equivalent quark helicity-dependent FF in single pion SIDIS, the DiFF GK 1 offers a unique lens into novel spin-dependent fragmentation. For instance, the twist-2 modulations observed in this study are enhanced by vector mesons created during fragmentation — a behavior predicted by phenomenological models. This study broadens our understanding of dihadron fragmentation, revealing new details about the flavor and charge dependence of hadronization.

Matousek, Gregory [Duke Univ., Durham, NC (United ↗
DOE OSTI2022

Deep Learning At Depth: Estimating subsurface parameters from geophysical monitoring data

Geophysical imaging techniques are a non-invasive way to image the subsurface and understand both subsurface solid (rock/soil) and fluid property distributions and their evolution in time. Inversions of the geophysical data, such as Electrical Resistance Tomography (ERT) data, are solved to estimate the subsurface property distributions, such as conductivity, and many inversion techniques smooth out sharp gradients in rock or fluid property distributions. Sharp gradients in subsurface properties tend to be present in situations with complex subsurface structures, which are common in many subsurface applications. We have successfully demonstrated that it is possible to inform, or constrain, inversions with neural networks trained on synthetic data with complex subsurface structures. Initial results suggest this process may be optimizable to yield property distributions that better represent the true property distributions than the same inversion process without the neural network constraint. Future work would optimize the neural network performance for this application and then apply the synthetic-data trained neural network to real data to understand the utility and performance of this technique for real data sets.

47 OTHER INSTRUMENTATION↗
DOE OSTI2020

Deep Learning for Intelligent Bubble Size Detection in the Spallation Neutron Source Visual Target

The Spallation Neutron Source (SNS) at Oak Ridge National Laboratory (ORNL) will undergo proton power upgrade (PPU), increasing the proton beam power from 1.4 MW to 2.8 MW. From 2.8 MW, 2.0 MW will go to the current First Target Station and the rest will go to the future Second Target Station (STS). The First Target Station uses a liquid mercury target that is contained in a 316L stainless steel vessel. The proton beam is pulsed at 60 Hz, with a pulse of about 0.7μs. When the proton beam hits the target, the intense energy deposition leads to a rapid rise in temperature in the mercury. This temperature rise creates pressure waves that propagate through the mercury and cause cavitation erosion. The power upgrade will cause stronger pressure waves that will further increase damage because of cavitation. Injecting small helium bubbles in the mercury has been an efficient method of mitigating the pressure wave at 1.4 MW. However, at higher power, additional mitigation is necessary. Therefore, the 2 MW target vessel will be equipped with swirl bubblers and an additional gas injection port near the nose to inject more gas in the target. To develop a gas injection strategy and design, flow visualization in water with a transparent prototypical target (“visual target”) was performed. Bubble sizes and their spatial distribution in the flow loop are crucial to understanding the effectiveness of the bubbles in mitigating pressure waves. Bubbles were generated in the visual target under varied conditions of input pressures with helium and air. Images were captured using a high-speed camera at varied frame rates at different positions away from the swirl bubbler and different depths in the flow loop under varying lighting conditions. Initially, methods such as circular Hough transforms were applied after a series of images processing to obtain a general distribution of bubble sizes. Bubbles smaller than 500 μm are preferred to effectively mitigate the effect of pressure waves, which demands an accurate bubble detection and sizing system. Intelligent detection and identification of bubble sizes alleviate misdetection and improves accuracies. Employing neural networks, intelligent detection of bubble sizes and their distribution was developed and provides a robust alternative to traditional techniques. Human intervention was employed to label in-focus and out-of-focus bubbles in the set of training images. An object detection network using a pretrained convolutional neural network was created that extracted the features from the training images. Data augmentation was used to improve network accuracy through a random transformation of the original data.

Rasheed, Fayaz↗
DOE OSTI2024

Quantum Computing for Energy-Related Applications

Growing interest in quantum computing and simulations have created opportunities for its deployment to improve processes pertaining to energy production, distribution, and consumption. While quantum computing is considered as a paradigm shift in our basic understanding of physical computation, effective implementation of quantum computing in energy applications also depends on progress and development in the dimensions of both quantum computing hardware and quantum computing algorithms. To fully address the status and future challenges of quantum information science (QIS) applied within the energy sector, in this presentation, we firstly summarize recent advancements on the applications of quantum computing to energy infrastructure and materials, complex energy system processes, advanced manufacturing, and energy system security. Then, we will demonstrate the results of quantum computing both on a simulator and a quantum device accessing from OLCF. Our first example is to use the variational quantum eigensolver (VQE) with a unitary coupled cluster with singles and doubles (UCCSD) ansatz to simulate a series of LixHyq molecules (q=-1, 0, +1). The obtained results showed that the quantum computing VQE-UCCSD is comparable to classical CCSD for small systems like LiH with respect to full configuration interaction (FCI). Targeting on CO2 capture application, our second example is to use VQE to quantify molecular vibrational energies and reaction pathways between CO2 and a simplified amine-based solvent model—NH3 to form H2NCOOH. This research showcases quantum computing applications in the study of CO2 capture reactions.

Duan, Yuhua↗
DOE OSTI2023

Collision-energy dependence of the Breit-Wheeler process in heavy-ion collisions and its application to nuclear charge radius measurements

The collision energy dependence of the cross section and the transverse momentum distribution of dielectrons from the Breit-Wheeler process in heavy-ion collisions are computed in the lowest-order QED and found to be sensitive to the nuclear charge distribution and the infrared divergence of the ultra-Lorentz-boosted Coulomb field. Within a given experimental kinematic acceptance, the cross section is found to increase while the pair transverse momentum ($\sqrt{\langle{p^2_T}\rangle}$) decreases with increasing beam energy. We demonstrate that the transverse-momentum component of Weizsäcker-Williams photons is due to the finite extent of the charge source and electric field component in the longitudinal direction. We further clarify the connection between the nuclear charge distribution and the kinematics of produced e + e - from the Breit-Wheeler process, and propose a criterion for the validity of the Breit-Wheeler process in relativistic heavy-ion collisions. Following this approach we demonstrate that the experimental measurements of the Breit-Wheeler process in ultrarelativistic heavy-ion collisions can be used to quantitatively constrain the nuclear charge radius. The extracted parameters show sensitivity to the impact parameter dependence, and can be used to study the initial-state and final-state effects in hadronic interactions.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2020

The Milky Way’s bulge star formation history as constrained from its bimodal chemical abundance distribution

ABSTRACT We conduct a quantitative analysis of the star formation history (SFH) of the Milky Way’s (MW) bulge by exploiting the constraining power of its stellar [Fe/H] and [Mg/Fe] distribution functions. Using Apache Point Observatory Galactic Evolution Experiment survey data, we confirm the previously established bimodal [Mg/Fe]–[Fe/H] distribution within 3 kpc of the inner Galaxy. To fit the chemical bimodal distribution, we use a simple but flexible star formation framework, which assumes two distinct stages of gas accretion and star formation, and systematically evaluate a wide multidimensional parameter space. We find that the data favour a three-phase SFH that consists of an initial starburst, followed by a rapid star formation quenching episode, and a lengthy, quiescent secular evolution phase. The metal-poor, high-α bulge stars ([Fe/H] < 0.0 and [Mg/Fe] > 0.15) are formed rapidly (<2 Gyr) during the early starburst. The density gap between the high- and low-α sequences is due to the quenching process. The metal-rich, low-α population ([Fe/H] > 0.0 and [Mg/Fe] < 0.15) then accumulates gradually through inefficient star formation during the secular phase. This is qualitatively consistent with the early SFH of the inner disc. Given this scenario, a notable fraction of young stars (age <5 Gyr) is expected to persist in the bulge. Combined with extragalactic observations, these results suggest that a rapid star formation quenching process is responsible for bimodal distributions in both the MW’s stellar populations and in the general galaxy population and thus plays a critical role in galaxy evolution.

Lian, Jianhui↗
DOE OSTI2024

KOMPASS-II: Compaction of Crushed salt for Safe Containment – Phase 2

Long-term stable sealing elements are a basic component in the safety concept for a possible repository for heat-emitting radioactive waste in rock salt. The sealing elements will be part of the closure concept for drifts and shafts. They will be made from a welldefinied crushed salt in employ a specific manufacturing process. The use of crushed salt as geotechnical barrier as required by the German Site Selection Act from 2017 /STA 17/ represents a paradigm change in the safety function of crushed salt, since this material was formerly only considered as stabilizing backfill for the host rock. The demonstration of the long-term stability and impermeability of crushed salt is crucial for its use as a geotechnical barrier. The KOMPASS-II project, is a follow-up of the KOMPASS-I project and continues the work with focus on improving the understanding of the thermal-hydraulic-mechanical (THM) coupled processes in crushed salt compaction with the objective to enhance the scientific competence for using crushed salt for the long-term isolation of high-level nuclear waste within rock salt repositories. The project strives for an adequate characterization of the compaction process and the essential influencing parameters, as well as a robust and reliable long-term prognosis using validated constitutive models. For this purpose, experimental studies on long-term compaction tests are combined with microstructural investigations and numerical modeling. The long-term compaction tests in this project focused on the effect of mean stress, deviatoric stress and temperature on the compaction behavior of crushed salt. A laboratory benchmark was performed identifying a variability in compaction behavior. Microstructural investigations were executed with the objective to characterize the influence of pre-compaction procedure, humidity content and grain size/grain size distribution on the overall compaction process of crushed salt with respect to the deformation mechanisms. The created database was used for benchmark calculations aiming for improvement and optimization of a large number of constitutive models available for crushed salt. The models were calibrated, and the improvement process was made visible applying the virtual demonstrator.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2020

Bremsstrahlung in GRMHD Models of Accreting Black Holes

Here, the role of bremsstrahlung in the emission from hot accretion flows around slowly accreting supermassive black holes is not thoroughly understood. In order to appraise the importance of bremsstrahlung relative to other radiative processes, we compute spectral energy distributions (SEDs) of accretion disks around slowly accreting supermassive black holes including synchrotron radiation, inverse Compton scattering, and bremsstrahlung. We compute SEDs for (i) four axisymmetric radiative general relativistic magnetohydrodynamics (RadGRMHD) simulations of 10 8 M ⊙ black holes with accretion rates between 10 -8 $\dot{M}$ Edd and 10 -5 $\dot{M}$ Edd , (ii) four axisymmetric RadGRMHD simulations of M87* with varying dimensionless spin a$_*$ and black hole mass, and (iii) a 3D GRMHD simulation scaled for Sgr A*. At 10 -8 $\dot{M}$ Edd , most of the luminosity is synchrotron radiation, while at 10 -5 $\dot{M}$ Edd the three radiative processes have similar luminosities. In most models, bremsstrahlung dominates the SED near 512 keV. In the M87* models, bremsstrahlung dominates this part of the SED if a$_*$ = 0.5, but inverse Compton scattering dominates if a$_*$ = 0.9375. Since scattering is more variable than bremsstrahlung, this result suggests that 512 keV variability could be a diagnostic of black hole spin. In the appendix, we compare some bremsstrahlung formulæ found in the literature.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2023

Ensemble models for circuit topology estimation, fault detection and classification in distribution systems

This paper presents a methodology for simultaneous fault detection, classification, and topology estimation for adaptive protection of distribution systems. The methodology estimates the probability of the occurrence of each one of these events by using a hybrid structure that combines three sub-systems, a convolutional neural network for topology estimation, a fault detection based on predictive residual analysis, and a standard support vector machine with probabilistic output for fault classification. The input to all these sub-systems is the local voltage and current measurements. A convolutional neural network uses these local measurements in the form of sequential data to extract features and estimate the topology conditions. The fault detector is constructed with a Bayesian stage (a multitask Gaussian process) that computes a predictive distribution (assumed to be Gaussian) of the residuals using the input. Since the distribution is known, these residuals can be transformed into a Standard distribution, whose values are then introduced into a one-class support vector machine. The structure allows using a one-class support vector machine without parameter cross-validation, so the fault detector is fully unsupervised. Finally, a support vector machine uses the input to perform the classification of the fault types. All three sub-systems can work in a parallel setup for both performance and computation efficiency. In conclusion, we test all three sub-systems included in the structure on a modified IEEE123 bus system, and we compare and evaluate the results with standard approaches.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2024

Measuring thermal profiles in high explosives using neural networks

We present a new method for calculating the temperature profile of high explosive (HE) material using a Convolutional Neural Network (CNN). To train/test the CNN, we have developed a hybrid experiment/simulation method for collecting acoustic and temperature data. We experimentally heat cylindrical containers of HE material until detonation/deflagration, where we continuously measure the acoustic bursts through the HE using multiple acoustic transducers lined around the exterior container circumference. However, measuring the temperature profile in the HE in an experiment would require inserting a large number of thermal probes, which would disrupt the heating process. Thus, we use two thermal probes, one at the HE center and one at the wall. We then use numerical simulation of the heating process to calculate the temperature distribution and correct the simulated temperatures based on the experimental center and wall temperatures. We calculate temperature errors on the order of 15 °C, which is ∼12% of the range of temperatures in the experiment. We also investigate how the algorithm’s accuracy is affected by the number of acoustic receivers used to collect each measurement and the resolution of the temperature prediction. This work provides a means of assessing the safety status of HE material, which cannot be achieved using existing temperature measurement methods. In addition, it has implications for a range of other applications where internal temperature profile measurements would provide critical information. These applications include detecting chemical reactions, observing thermodynamic processes such as combustion, monitoring metal or plastic casting, determining the energy density in thermal storage capsules, and identifying abnormal battery operations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Measuring thermal profiles in high explosives using neural networks

We present a new method for calculating the temperature profile of high explosive (HE) material using a Convolutional Neural Network (CNN). To train/test the CNN, we have developed a hybrid experiment/simulation method for collecting acoustic and temperature data. We experimentally heat cylindrical containers of HE material until detonation/deflagration, where we continuously measure the acoustic bursts through the HE using multiple acoustic transducers lined around the exterior container circumference. However, measuring the temperature profile in the HE in an experiment would require inserting a large number of thermal probes, which would disrupt the heating process. Thus, we use two thermal probes, one at the HE center and one at the wall. We then use numerical simulation of the heating process to calculate the temperature distribution and correct the simulated temperatures based on the experimental center and wall temperatures. We calculate temperature errors on the order of 15 °C, which is ~12% of the range of temperatures in the experiment. We also investigate how the algorithm’s accuracy is affected by the number of acoustic receivers used to collect each measurement and the resolution of the temperature prediction. This work provides a means of assessing the safety status of HE material, which cannot be achieved using existing temperature measurement methods. In addition, it has implications for a range of other applications where internal temperature profile measurements would provide critical information. These applications include detecting chemical reactions, observing thermodynamic processes such as combustion, monitoring metal or plastic casting, determining the energy density in thermal storage capsules, and identifying abnormal battery operations.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Combined Data and Deep Learning Model Uncertainties: An Application to the Measurement of Solid Fuel Regression Rate

In complex physical process characterization, such as the measurement of the regression rate for solid hybrid rocket fuels, where both the observation data and the model used have uncertainties originating from multiple sources, combining these in a systematic way for quantities of interest (QoI) remains a challenge. In this paper, we present a forward propagation uncertainty quantification (UQ) process to produce a probabilistic distribution for the observed regression rate r. We characterized two input data uncertainty sources from the experiment (the distortion from the camera U c and the non-zero-angle fuel placement U Y ), the prediction and model form uncertainty from the deep neural network (U m ), as well as the variability from the manually segmented images used for training it (U s ). Here, we conducted seven case studies on combinations of these uncertainty sources with the model form uncertainty. The main contribution of this paper is the investigation and inclusion of the experimental image data uncertainties involved, and how to include them in a workflow when the QoI is the result of multiple sequential processes.

42 ENGINEERING↗
DOE OSTI2024

New physical processes for extracting GPDs with a better sensitivity to partonic structure

We introduce a new type of exclusive processes for a better study of generalized parton distributions (GPDs), which we refer to as single-diffractive hard exclusive processes (SDHEPs). We advocate a two-stage framework for picturing SDHEPs based on the separation of scales, which gives a clear description both kinematically and dynamically. We examine the sensitivity of the SDHEP to the parton momentum fraction x -dependence of GPDs, and demonstrate it quantitatively with two specific processes that can be readily measured at J-PARC or AMBER using a pion beam and at JLab using a photon beam, respectively. Both processes are capable of providing enhanced sensitivity to the x -dependence, overcoming the problem of shadow GPDs, and disentangling different types of GPDs with various spin asymmetries.

Qiu, Jianwei↗
DOE OSTI2025

High dopant activation in arsenic doped single-crystal CdTe thin films: Insights from MBE growth and rapid thermal processing

Single-crystal model systems are valuable tools to investigate fundamental material properties. In this work, we use molecular beam epitaxy to deposit in situ arsenic (As) doped single-crystal CdTe films on large area Si substrates to better understand As doping for photovoltaic applications. We found that As incorporation is highly temperature dependent: a substrate temperature difference of 50 °C can lead to several orders of magnitude difference in As concentration. Cd overpressure during in situ doping may limit out-diffusion of As but decrease As incorporation, especially at lower growth temperatures. Carrier concentrations greater than 10 16 cm −3 can be achieved with or without Cd overpressure when annealed at temperatures above 500 °C. However, unlike the low (∼1% to 5%) dopant activation commonly observed in polycrystalline CdTe, our films achieve significantly higher activation ratios—exceeding 50%, and in some cases approaching 80%. These values are consistent with or exceed prior reports in single-crystal CdTe systems. In addition to as-deposited arsenic concentrations, we also consider arsenic distribution after different rapid thermal processing temperatures. We propose a detailed definition and description of how arsenic incorporation is considered and calculated. Due to carrier concentration saturation, As incorporation also needs to be controlled to average levels of 10 17 cm −3 to achieve high activation. These findings suggest that higher annealing temperature regimes may be beneficial to polycrystalline CdTe based PV devices.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Magnetic Field Sensing via Acoustic Sensing Fiber with Metglas® 2605SC Cladding Wires

Magnetic field sensing has the potential to become necessary as a critical tool for long-term subsurface geophysical monitoring. The success of distributed fiber optic sensing for geophysical characterization provides a template for the development of next generation downhole magnetic sensors. In this study, Sentek Instrument’s picoDAS is coupled with a multi-material single mode optical fiber with Metglas® 2605SC cladding wire inclusions for magnetic field detection. The response of acoustic sensing fibers with one and two Metglas® 2605SC cladding wires was evaluated upon exposure to lateral AC magnetic fields. An improved response was demonstrated for a sensing fiber with in-cladding wire following thermal magnetic annealing (~400 °C) under a constant static transverse magnetic field (~200 μT). A minimal detectable magnetic field of ~500 nT was confirmed for a sensing fiber with two 10 μm cladding wires. The successful demonstration of a magnetic field sensing fiber with Metglas® cladding wires fabricated via traditional draw processes sets the stage for distributed measurements and joint inversion as a compliment to distributed fiber optic acoustic sensors.

Dejneka, Zach (ORCID:0000000179415708)↗
DOE OSTI2022

r -process nucleosynthesis and kilonovae from hypermassive neutron star post-merger remnants

ABSTRACT We investigate r-process nucleosynthesis and kilonova emission resulting from binary neutron star (BNS) mergers based on a three-dimensional (3D) general-relativistic magnetohydrodynamic (GRMHD) simulation of a hypermassive neutron star (HMNS) remnant. The simulation includes a microphysical finite-temperature equation of state (EOS) and neutrino emission and absorption effects via a leakage scheme. We track the thermodynamic properties of the ejecta using Lagrangian tracer particles and determine its composition using the nuclear reaction network SkyNet. We investigate the impact of neutrinos on the nucleosynthetic yields by varying the neutrino luminosities during post-processing. The ejecta show a broad distribution with respect to their electron fraction Ye, peaking between ∼0.25–0.4 depending on the neutrino luminosity employed. We find that the resulting r-process abundance patterns differ from solar, with no significant production of material beyond the second r-process peak when using luminosities recorded by the tracer particles. We also map the HMNS outflows to the radiation hydrodynamics code SNEC and predict the evolution of the bolometric luminosity as well as broadband light curves of the kilonova. The bolometric light curve peaks on the timescale of a day and the brightest emission is seen in the infrared bands. This is the first direct calculation of the r-process yields and kilonova signal expected from HMNS winds based on 3D GRMHD simulations. For longer-lived remnants, these winds may be the dominant ejecta component producing the kilonova emission.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2020

Feasibility of Process Intensification of Water-Gas Shift Reaction Using a Microreactor with Integrated Cooling

Hydrogen is an increasingly attractive low-carbon energy carrier for a variety of stationary and mobile applications. The water-gas shift (WGS) reaction is a key processing step used for large-scale hydrogen production via the steam methane reforming process. However, the thermodynamics and kinetics of the reaction are such that standard two-stage adiabatic reactors used in these systems are large, increasing catalyst volume and cost. To intensify the process and realize the economical distributed production of hydrogen, adiabatic WGS microreactors with integrated cooling directly regulate the reaction temperature via integrated coolant channels to promote higher conversion within a smaller reactor volume. This study investigates the conversion efficiency of a single WGS microchannel operating under such cooling conditions. A COMSOL Multiphysics model is developed and validated with isothermal experimental data from the literature. The model is then used to evaluate improvements in conversion efficiency when the reaction is cooled via a specified wall temperature profile. Lastly, the model is modified to include cooling channels with a secondary fluid that can practically achieve a similar conversion profile as the specified wall temperature profile previously applied. Initial results show that reactor conversion can be significantly increased by the inclusion of appropriate cooling and that there is a potential for the recovery of energy from the reaction stream that can be used for other applications within the overall process.

energy recovery↗
DOE OSTI2024

Distributed Multi-GPU Community Detection on Exascale Computing Platforms

Community detection is a fundamental operation in graph mining, and by uncovering hidden structures and patterns within complex systems it helps solve fundamental problems pertaining to social networks, such as information diffusion, epidemics, and recommender systems. Scaling graph algorithms for massive networks becomes challenging on modern distributed-memory multi-GPU (Graphics Processing Unit) systems due to limitations such as irregular memory access patterns, load imbalances, higher communication-computation ratios, and cross-platform support. We present a novel algorithm HiPDPL-GPU (distributed parallel Louvain) to address these challenges. We conduct experiments involving different partitioning techniques to achieve optimized performance of HiPDPL-GPU on the two largest supercomputers: Frontier and Summit. Remarkably, HiPDPL-GPU processes a graph with 4.2 billion edges in less than 3 minutes using 1024 GPUs. Qualitatively performance of HiPDPL-GPU is similar or better compared to other state-of-the-art CPU- and GPU-based implementations. While prior GPU implementations have predominantly employed CUDA, our first-of-its-kind implementation for community detection is cross-platform, accommodating both AMD and NVIDIA GPUs.

graph algorithms, high performance comptuing↗