Search NASA⌕ Search

SEARCH · Search NASA

Results for “process and scheduling”

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 343 records · Page 19
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

The Galaxy platform for accessible, reproducible, and collaborative data analyses: 2024 update

Galaxy (https://galaxyproject.org) is deployed globally, predominantly through free-to-use services, supporting user-driven research that broadens in scope each year. Users are attracted to public Galaxy services by platform stability, tool and reference dataset diversity, training, support and integration, which enables complex, reproducible, shareable data analysis. Applying the principles of user experience design (UXD), has driven improvements in accessibility, tool discoverability through Galaxy Labs/subdomains, and a redesigned Galaxy ToolShed. Galaxy tool capabilities are progressing in two strategic directions: integrating general purpose graphical processing units (GPGPU) access for cutting-edge methods, and licensed tool support. Engagement with global research consortia is being increased by developing more workflows in Galaxy and by resourcing the public Galaxy services to run them. The Galaxy Training Network (GTN) portfolio has grown in both size, and accessibility, through learning paths and direct integration with Galaxy tools that feature in training courses. Code development continues in line with the Galaxy Project roadmap, with improvements to job scheduling and the user interface. Environmental impact assessment is also helping engage users and developers, reminding them of their role in sustainability, by displaying estimated CO 2 emissions generated by each Galaxy job.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Experiences Detecting Defective Hardware in Exascale Supercomputers

In May 2022, the newest supercomputer to top the TOP 500 list was Frontier at Oak Ridge National Laboratory, demonstrating the capability of computing more than 1.1 quintillion (1018) floating-point calculations every second. Driving this ground-breaking rate of computing is Frontier’s more than 37,000 graphics processing units (GPUs) and 9,408 central processing units (CPUs). In total, Frontier contains more than 60 million parts. At this scale, the smallest margin of error may generate hundreds of hardware errors across the system. These errors are capable of directly hindering world-class science performed on Frontier if not found. In this work, we describe and evaluate two strategies for finding hardware-level faults in Frontier’s 9,408 compute nodes. There are two strategies developed: the first uses the Slurm scheduler to scavenge available compute time to run the node screen, the second builds upon the lessons learned in the first strategy and enforces a weekly screen of each node. Using June 2023 as a case study, we find that the first scheduling strategy consumed more than ten times the resources as the second scheduling strategy, but successfully detected five hardware defects in Frontier. We summarize the lessons learned while developing and running a node screen on the world’s first exascale supercomputer.

Hagerty, Nick↗
DOE OSTI2023

EV charging site day-ahead load prediction in a synthetic environment for RL based grid-informed charging

Ensuring grid health in the face of increasing demand for power is an emerging challenge especially due to transportation electrification. A free market approach to influencing electric vehicle (EV) load through grid-informed hourly dynamic pricing is introduced in this work. The setting of charging price is done by a reinforcement learning (RL) agent that learns the complicated dynamics by interacting with a synthetic environment. This synthetic environment is a combination of distribution feeder simulation, EV charger user behavior dynamics, and EV charging simulation. A key module in this synthetic environment involves obtaining the day-ahead charging profile of EV charging stations based on real-world past data. The day-ahead prediction is also useful in other traditional optimizations related to EV charge scheduling. The proposed approach involves using EV charging data from two different past time horizons – one to determine the shape of the daily profile and the other to determine a scaling value to capture actual energy consumption. Real-world charging data over many years from the ACN charging network has been used to demonstrate the ability to predict the day-ahead profile with only charge session data. Both Python and MATLAB have been used for data cleaning, processing, analysis, and prediction.

Suryanarayana, Harish↗
DOE OSTI2020

Y-12 Groundwater Protection Program Groundwater and Surface Water Sampling and Analysis Plan (CY 2021)

This plan provides a description of the groundwater and surface water quality monitoring activities planned for calendar year (CY) 2021 at the U.S. Department of Energy Y-12 National Security Complex (Y-12) that will be managed by the Y-12 Groundwater Protection Program (GWPP). Groundwater and surface water monitoring is performed by the GWPP. Groundwater and surface water monitoring will be performed in three hydrogeologic regimes at Y-12: the Bear Creek Hydrogeologic Regime (Bear Creek Regime), the Upper East Fork Poplar Creek Hydrogeologic Regime (East Fork Regime), and the Chestnut Ridge Hydrogeologic Regime (Chestnut Ridge Regime). The Bear Creek and East Fork regimes are located in Bear Creek Valley and the Chestnut Ridge Regime is located south of Y-12. Additional surface water monitoring will be performed north of Pine Ridge along the boundary of the Oak Ridge Reservation. The following sections of this report provide details regarding the CY 2021 groundwater and surface water monitoring activities. Section 2 describes the monitoring locations in each regime and the processes used to select the sampling locations. A description of the field measurements and laboratory analytes is provided in Section 3. Sample collection methods and procedures are described in Section 4, and Section 5 lists the documents cited for more detailed operational and technical information. The narrative sections of the report reference several appendices. Figures (maps and diagrams) and tables (excluding a data summary table presented in Section 4) are in Appendix A and Appendix B, respectively. Groundwater Monitoring Schedules (when issued throughout CY 2021) will be inserted in Appendix C, and addenda to this plan (if issued) will be inserted in Appendix D. Laboratory requirements (bottle lists, holding times, etc.) are provided in Appendix E, and an approved Waste Management Plan is provided in Appendix F. Modifications to the CY 2021 monitoring program may be necessary during implementation. Changes in programmatic requirements may alter the analytes specified for selected monitoring wells or may add or remove wells from the planned monitoring network. Each modification to the monitoring program will be approved by the Y-12 GWPP manager and documented as an addendum to this sampling and analysis plan.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Forward Physics Facility

High-energy collisions at the High-Luminosity Large Hadron Collider will produce a large number of particles along the beam collision axis, outside of the acceptance of existing LHC experiments. The proposed Forward Physics Facility (FPF), to be located 617--682 meters from the ATLAS interaction point along the beam collision axis and shielded by concrete and rock, will host a suite of experiments to probe Standard Model processes and search for physics beyond the Standard Model. The FPF is supported by a rapidly growing user community of 400 physicists, and four pathfinder detectors that are currently operating at the LHC have already yielded promising results, including the first direct detection of collider neutrinos. Here we provide an Executive Summary, and then briefly summarize the status of the FPF, including its rich science case, the facility itself, the proposed FPF experiments and their technical readiness, the preliminary schedule and budget profile, and plans for international participation. Submitted to the DOE/NSF HEPAP Particle Physics Projects Prioritization Panel~(P5)

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2021

Dynamic Distribution of High-Rate Data Processing from CERN to Remote HPC Data Centers

The prompt reconstruction of the data recorded from the Large Hadron Collider (LHC) detectors has always been addressed by dedicated resources at the CERN Tier-0. Such workloads come in spikes due to the nature of the operation of the accelerator and in special high load occasions experiments have commissioned methods to distribute (spill-over) a fraction of the load to sites outside CERN. The present work demonstrates a new way of supporting the Tier-0 environment by provisioning resources elastically for such spilled-over workflows onto the Piz Daint Supercomputer at CSCS. Furthermore, this is implemented using containers, tuning the existing batch scheduler and reinforcing the scratch file system, while still using standard Grid middleware. ATLAS, CMS and CSCS have jointly run selected prompt data reconstruction on up to several thousand cores on Piz Daint into a shared environment, thereby probing the viability of the CSCS high performance computer site as on demand extension of the CERN Tier-0, which could play a role in addressing the future LHC computing challenges for the high luminosity LHC.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

A Generation-Storage Coordination Dispatch Strategy for Power System Based on Causal Reinforcement Learning

In the backdrop of global energy transformation, power systems integrating high proportions of renewable energy sources are facing unprecedented challenges in operational stability and dispatch efficiency. To address these challenges, this study introduces a generation-storage coordination real-time dispatch strategy based on Causal Power System Dynamic Reinforcement Learning (CPSDRL). Diverging from traditional reinforcement learning approaches, CPSDRL innovatively incorporates causal inference within the state prediction model - the crux of model-based reinforcement learning - thereby establishing the Power Causal Dynamic Model (PCDM). Assisted by the prior knowledge of power systems, the model significantly enhances prediction accuracy and reliability through a two-stage training process. Utilizing PCDM, this study further applies a direct policy search algorithm to optimize the real-time dispatch strategy. Experimental results indicate that the proposed method improves the stability of generation-storage coordination real-time dispatch and exhibits competitive advantages in sample efficiency and computational speed, compared to traditional model-based and model-free reinforcement learning algorithms. This method is expected to enhance the practicality and adaptability of causal reinforcement learning techniques in power system scheduling and control.

causal reinforcement learning↗
DOE OSTI2026

Used Nuclear Fuel Management Using the Next Generation System Analysis Model

The U.S. Department of Energy (DOE) is leading the National effort to manage the back end of the nuclear fuel cycle, encompassing the safe transportation, storage/staging, and/or eventual disposal of used nuclear fuel (UNF) and high-level radioactive waste. The Next Generation System Analysis Model (NGSAM) is DOE’s discrete-event, agent-based simulation tool designed to model the full life cycle of UNF from reactor discharge to final disposal. NGSAM supports the DOE Office of Spent Fuel and High-Level Waste Disposition by enabling a detailed, scenario-based analysis of logistics, infrastructure, and shipping strategies. NGSAM replaces legacy models with a modern, flexible platform built on Repast Simphony and enhanced by the Process Analysis Tool. NGSAM simulates the movement and interaction of individual fuel assemblies with system components such as canisters, casks, railcars, and facilities. The model integrates with the Java Transportation Operations Model to plan and execute transportation scenarios, supporting both constrained and unconstrained resource allocation. Key features include customizable allocation and acceptance algorithms, detailed facility-level operations, and a Quick Edit tool for rapid scenario adjustments. NGSAM supports multimodal transportation modeling (e.g. rail, road, barge) and provides comprehensive cost, schedule, and infrastructure data. NGSAM utilizes data from sources such as DOE’s STANDARDS UNF database and DOE’s Stakeholder Tool for Assessing Radioactive Transportation, while also allowing user-defined inputs for scenario customization. NGSAM enables stakeholders to evaluate complex UNF management strategies, assess system performance under varying assumptions, and inform decision making for future infrastructure investments. Its modular architecture and integration with other Integrated Waste Management System tools make it a critical asset for planning the safe and efficient disposition of the Nation’s growing UNF inventory.

Craig, Brian [Argonne National Laboratory (ANL)]↗
DOE OSTI2025

Detection of visible-wavelength aurora on Mars

Mars hosts various auroral processes despite the planet’s tenuous atmosphere and lack of a global magnetic field. To date, all aurora observations have been at ultraviolet wavelengths from orbit. We describe the discovery of green visible-wavelength aurora, originating from the atomic oxygen line at 557.7 nanometers, detected with the SuperCam and Mastcam-Z instruments on the Mars 2020 Perseverance rover. Near–real-time simulations of a Mars-directed coronal mass ejection (CME) provided sufficient lead-time to schedule an observation with the rover. The emission was observed 3 days after the CME eruption, suggesting that the aurora was induced by particles accelerated by the moving shock front. To our knowledge, detection of aurora from a planetary surface other than Earth has never been reported, nor has visible aurora been observed at Mars. This detection demonstrates that auroral forecasting at Mars is possible, and that during events with higher particle precipitation, or under less dusty atmospheric conditions, aurorae will be visible to future astronauts.

Science & Technology - Other Topics↗
DOE OSTI2025

Seismoacoustic Monitoring of Pyroprocessing Equipment

With the projected increase in world nuclear capacity comes the hurdle of spent nuclear fuel. Pyroprocessing is one method to process irradiated fuel by using high temperatures and electrochemical steps to separate radioactive components. High temperatures, high radiation levels, and equipment confined to a heavily shielded hot cell are a few of the challenges introduced regarding safeguards for pyroprocessing. The first stage of the process utilizes an element chopper to cut irradiated fuel into smaller pieces, offering the potential to identify equipment operation through seismoacoustic monitoring. Eleven seismic and infrasound sensors were deployed at various locations and distances near a pyroprocessing hot cell to evaluate signals emitted from the chopper. Signals were processed using short-time Fourier transforms and manually scanned to locate chopping events. Operator logs were then collected for a ground truth comparison. Results indicate that both infrasound and seismic signals offer the ability to accurately measure this stage of pyroprocessing, though seismic signals are more prominent. Sensor locations are also evaluated to determine where the signal is detectable. This type of monitoring offers a method to aid in safeguards and proliferation detection by validating a pyroprocessing facility’s schedule and identifying activity that does not align with records.

11 - NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2020

DOE Hanford Tank Side Cesium Removal System (TSCR) Update: Fabrication and Factory Acceptance Testing - 20300

The Tank Side Cesium Removal (TSCR) System is a modular, quickly-deployable, and passively safe system for efficiently removing solids and cesium from alkaline liquid waste raffinate created by fuel processing on the Hanford site. This paper provides an overview of the project with particular emphasis on completion of fabrication and factory acceptance testing (FAT). All TSCR System components are inside three enclosures known as the Process Enclosure, Ancillary Enclosure, and Control Enclosure. The Process Enclosure is seismic design category II (SDC-II) and built to withstand the rigors of transportation and lifting, high winds, volcanic ash fall, and heavy precipitation (snow) over a temperature range of -25 to 115 deg. F. The Process Enclosure and Control Enclosure were fabricated in AVANTech's Columbia, SC facility, while the Ancillary Enclosure was fabricated in AVANTech's Richland, WA facility. The design and preliminary documented safety analysis were completed in December 2019, with all work being performed in accordance with ASME NQA-1 quality assurance requirements. The Process Enclosure, Ion Exchange Columns (IXCs), and several other TSCR components have a nuclear safety function and are designated as safety significant structures, systems, and components (SS-SSCs). The NQA-1 Commercial Grade Dedication (CGD) process was used to identify critical characteristics in order to accept materials and items used for fabrication of several safety significant TSCR components. This paper also addresses unique aspects and challenges of implementing CGD in DOE facilities and systems. After fabrication and preliminary testing, the Process and Control Enclosures were shipped to AVANTech's Richland facility where the complete TSCR System was setup and connected in a represented layout and configuration that will be used on the Hanford site. Test tanks and pumps were put in place and connected to the TSCR Process Enclosure in a manner that mimicked interfacing Double Shell Tanks in the 241-AP Tank Farm. The TSCR System was then tested/operated under conditions equivalent to those planned for onsite operation. Testing was separated into five separate phases as defined below: - Phase 1: Confirmation of TSCR System readiness for the FAT; - Phase 2: Verification of TSCR System operation under standard conditions; - Phase 3: Demonstration of TSCR System operation under non-standard conditions; - Phase 4: Demonstration of ion exchange media removal from an IXC; and - Phase 5: Completion of TSCR System lift plans and site mobilization. The Hanford Tank Operations Contractor, Washington River Protection Solutions (WRPS), participated in all phases of testing. Testing at AVANTech's Richland Facility allowed WRPS personnel to familiarize themselves with TSCR operations in a radiologically clean facility prior to onsite mobilization. Engineering and operations personnel gained valuable experience that allowed them to better refine onsite operational protocols and procedures - including handling and movement of IXCs with a forklift. This paper describes the technologies used by the TSCR System as well as how these technologies are deployed and operated to advance the Hanford mission of safely, efficiently, and effectively treating tank waste. It also provides detailed information on fabrication and FAT, including operational data from simulated tank waste runs. A project schedule for WRPS' planned construction acceptance testing, operational acceptance testing, and actual tank waste treatment is provided. This presentation should be of great interest to parties responsible for the design, testing, and mobilization of modular NQA-1 technologies for the treatment of raffinate and similar liquid wastes containing high concentrations of radioactive cesium. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2025

Controls Status for Fermilab’s PIP-II Project

Testing of Cryomodule first articles at PIP-II's on-site Cryomodule Test Facility (CMTF), where EPICS based controls are already being used, has been invaluable to the project. In addition to validating hardware and controls components, it has provided opportunities to build buy-in and illicit meaningful feedback early-on. A device naming convention has also been significantly refined with an internal working group and the PIP-II integration team. Compressed air and Process Control Water (PCW) systems have been installed, and operations has been engaged in Phoebus display development in preparation for a transition to operations for these systems by 2026. Cryoplant hardware has been put in place, and collaboration between controls and PIP-IIs cryo group for IOC and display development is well underway for commissioning of cryoplant systems to begin early 2026. The project has also received beneficial occupancy of the High-Bay Building in October 2025, keeping us on schedule for Warm Front End commissioning to begin in 2027.

Crisp, Dan [Fermilab]↗
DOE OSTI2023

Fabrication, Modeling, and Testing of a Prototype for Particle Thermal Energy Storage Containment

Increasing penetration of variable renewable energy resources requires the deployment of energy storage at a range of durations. Long-duration energy storage (LDES) technologies will fulfill the need to firm variable renewable energy resource output throughout the year. Conventional electrochemical batteries (e.g., lithium-ion) are uneconomical in this role due to high energy capacity costs. Thermal energy storage (TES) is one promising technology for LDES applications because of its siting flexibility and ease of scaling. Particle-based TES systems use low-cost solid particles that have higher temperature limits than the molten salts used in traditional concentrated solar power systems. A key component in particle-based TES systems is the containment silo for the high-temperature (> 1100 degrees C) particles. This study combined experimental testing and computational modeling methods to design and characterize the performance of a particle containment silo for LDES applications. A containment silo prototype was built at a laboratory scale and used to validate a congruent transient finite element analysis (FEA) model. The validation compared the actual and predicted temperature profile through the prototype over six days as the particles cooled from their initial temperature. The performance of a commercial scale (> 5 GWhth) was then characterized using the validated model. The transient FEA model was subject to several charge-discharge cycles to mimic a possible operating schedule. The commercial-scale model predicted a storage efficiency in excess of 95% after five days of storage with a design storage temperature of 1200 degrees C. Insulation material and concrete temperature limits were considered as well. The validation of the methodology means the FEA model can simulate a range of scenarios for future applications. This work supports the development of a promising LDES technology with implications for grid-scale electrical energy storage, but also for thermal energy storage for industrial process heating applications.

ENERGY STORAGE↗
DOE OSTI2026

Multi-Agent Control Planes for Quantum Networks: A Scalable Architecture for Autonomous Quantum Internet Management

Quantum networks are expected to enable distributed quantum computing, secure communication, and global entanglement distribution. However, operating such networks presents significant challenges, including stochastic quantum processes, fragile entanglement resources, dynamic topology, and cross-layer control requirements. Current quantum network control architectures largely rely on centralized or hierarchical controllers inspired by classical software-defined networking (SDN). While effective for small testbeds, these approaches face scalability, latency, and reliability limitations as quantum networks grow. This paper proposes a multi-agent control plane architecture for quantum networks. In this design, intelligent software agents operate at quantum nodes, repeaters, and orchestration layers, collectively managing entanglement generation, routing, purification, and scheduling. The distributed intelligence of the agent system allows the network to adapt dynamically to quantum hardware variability and environmental noise. We argue that multi-agent systems provide significant advantages over centralized control approaches, including scalability, resilience, local autonomy, and real-time adaptation. The paper discusses architectural design principles, agent coordination mechanisms, and research challenges in deploying multi-agent control planes for the emerging quantum Internet.

Alnajjar, Anees [ORNL] (ORCID:0000000237101601)↗
DOE OSTI2022

Utilization of the LMP Methodology in Support of the VTR Conceptual Safety Design Report

The Versatile Test Reactor (VTR) is a fast spectrum test reactor currently being developed in the United States under the direction of the US Department of Energy (DOE), Office of Nuclear Energy. The VTR is utilizing a risk-informed performance-based (RIPB) approach for design support and authorization by the DOE, derived from recent efforts by the US industry led Licensing Modernization Project (LMP). This document contains an overview of the implementation of the LMP approach in support of the VTR Conceptual Safety Design Report (CSDR). The work reported here is the result of studies supporting a VTR conceptual design, cost, and schedule estimate for DOE-NE to make a decision on procurement. As such, it is preliminary. The VTR RIPB authorization approach utilizes information from the probabilistic risk assessment (PRA), coupled with deterministic analyses, to aid in decision-making regarding the identification and categorization of safety basis events (SBEs), the classification of structures, systems, and components (SSCs), and the evaluation of defense-in-depth (DID) adequacy. As part of initial reactor design efforts, a VTR conceptual design PRA was developed to support the RIPB process, which focused on at-power internal events, with scoping analyses for seismic and sodium fire hazards. In addition to supporting numerous design studies, preliminary results from the RIPB approach and the VTR conceptual design PRA were utilized as the basis of the VTR CSDR. The initial identification and categorization of SBEs, SSC classification, and DID evaluation were contained within the CSDR, which was submitted to DOE in 2019 as part of the CD-1 submittal package. Following review, DOE approved the CSDR in April 2020 and the CD-1 package in late 2020. Valuable experience was gained through the implementation of the RIPB approach for design and authorization during the VTR conceptual design phase, which is summarized in this document. To the extent possible, this experience has been shared with the advanced reactor industry, through publications and participation in licensing tabletops, in addition to informing DOE:NE advanced reactor regulatory development efforts. Furthermore, the approval of the CSDR by the DOE as part of CD-1 represents a significant milestone in the use of RIPB approaches for advanced reactor licensing.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2024

Leverage modern artificial intelligence (AI) enabled systems for waste reduction

Manufacturing industries continue to face challenges in reducing waste, as upstream strategies such as source reduction and product redesign require a deeper understanding of processes compared to conventional recycling methods. Recent advancements in artificial intelligence (AI) and machine learning (ML) have opened new opportunities to integrate modern computational techniques with traditional waste minimization strategies. This paper explores AI-enabled approaches for product redesign, source reduction, and recycling that can significantly reduce waste generation while improving efficiency and sustainability. AI-driven material substitution and lightweighting in product design enable discovery of novel materials with optimized properties, reducing waste without compromising performance. Reinforcement learning models optimize process parameters, raw material specifications, and machine sequencing to minimize production losses, while Industrial Internet of Things (IIoT) systems paired with AI analytics enhance real-time waste tracking, predictive maintenance, and quality inspection. Furthermore, AI-based demand forecasting and production planning reduce overproduction and excess inventory, as demonstrated in industrial applications. In recycling, ML-powered pattern recognition and robotic sorting technologies achieve higher accuracy in waste segregation, directly improving recycling efficiency. Complementary solutions such as smart bins and AI-enabled waste pickup scheduling optimize collection logistics, reducing both costs and emissions. Although implementation requires upfront investment in infrastructure and training, the long-term benefits include higher material efficiency, reduced waste, improved product quality, and stronger sustainability outcomes across the supply chain. By leveraging AI-enabled systems, manufacturers can align waste minimization efforts with circular economy principles, creating scalable solutions for both industry and society.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2023

Development of Novel Ferritic-Martensitic Steels with Superior Creep Properties for Power Plant Applications

A creep resistant martensitic steel, CPJ-7, was developed with an operating temperature approaching 650°C. Subsequently, another creep resistant martensitic steel, JMP, was designed with the potential to operate at, or slightly above, 650°C. This report describes the development of these alloys from early iterations of CPJ steels to the final CPJ-7 formulation as well as the more heat resistant JMP steel formulation. The design originated from computational modeling for phase stability and precipitate strengthening using fifteen constituent elements. Approximately forty heats of CPJ and ten of JMP, each weighing ~7 kg, were vacuum induction melted. A computationally optimized heat treatment schedule was developed to homogenize the ingots prior to hot forging and rolling prior to final normalization and tempering. Overall, wrought and cast versions of CPJ-7 present superior creep properties when compared to wrought and cast versions of COST alloys for steam turbine and wrought and cast versions of P91/92 for boiler applications. For instance, the Larson Miller Parameter curve for CPJ-7 at 650°C almost coincides with that of COST E at 620°C. The prolonged creep life was attributed to slowing down the process of the destabilization of the MX and M 23 C 6 precipitates at 650°C. On the other hand, the cast version of CPJ-7 also revealed superior mechanical performance, well above commercially available cast 9% Cr martensitic steel or derivatives, especially for creep life at 650°C. The casting process employed slow cooling to simulate the conditions of a thick wall full-size steam turbine casing but utilized a separate homogenization step prior to final normalization and tempering. To advance the development of CPJ-7 for commercial applications, a process was used to scale up the production of the alloy using vacuum induction melting (VIM) and electroslag remelting (ESR), which underlined the importance of melt processing control of intentionally designed minor and trace elements in these advanced alloys. Following the work on CPJ-7, the JMP steels were designed with higher Co for increased solid solution strengthening, Si for oxidation resistance and increased W (with low Mo content) for matrix strength and stability as well as solid solution strengthening. The JMP steels showed increases in creep life compared to CPJ-7 between 118 to 150% at 650°C for testing at various stresses between 138 MPa and 207 MPa. On a Larson-Miller plot, the performance of the JMP steels surpasses that of state-of-the-art MARBN and other MARBN-type steels. The influence of various elements within the composition of the alloys on the microstructure and mechanical properties are discussed. This report presents approximately 420,000 h of in-house creep testing, the equivalent of almost 50 years of cumulative creep tests.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Accelerating Transformer-based Deep Learning Models on FPGAs using Column Balanced Block Pruning

Although Transformer-based language representations achieve state-of-the-art accuracy on various natural language processing (NLP) tasks, the large model size has been challenging the resource constrained computing platforms. Weight pruning, as a popular and effective technique in reducing the number of weight parameters and accelerating the Transformer, has been investigated on GPUs. However, the Transformer acceleration using weight pruning on field-programmable gate array (FPGAs) remains unexplored. This paper investigates the column balanced block-wise pruning on Transformer and designs an FPGA acceleration engine to customize the balanced blockwise matrix multiplication. We implement the Transformer model with proper hardware scheduling, and the experiments show that the Transformer inference on FPGA achieves 10.35 ms latency with the batch size of 32, which is 10.96 × speed up comparing to CPU platform and 2.08 × speed up comparing to GPU platform.

Peng, Hongwu↗