SEARCH · Engineering Papers

Results for “Computer Operations and Hardware”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 487 records · Page 27

All sources NASA NTRS DOE OSTI

DOE OSTI2026

Energy Systems Integration Facility Stewardship Summary: Fiscal Year 2025

A summary of NLR's stewardship of the nationally unique Energy Systems Integration Facility (ESIF) highlighting performance metrics, capability upgrades, and examples of R&D impact. 2025 brought a national focus on energy and ESIF is meeting the moment. All eyes are on data centers and domestic manufacturing and bringing the benefits of artificial intelligence to power system planning and operations. In step with national priorities, ESIF is building out capabilities that advance secure, reliable, and affordable power. ESIF hosted 190 multidisciplinary research projects, 855 high-performance computer users, and collaborated with 81 partners from industry, academia, research, and federal agencies. These research projects resulted in an AI method for detecting high-impedance faults with 90% accuracy, a grid controls demonstration in Connecticut, power quality validation of CorePower's flagship inductor, and a cybersecurity assessment of potential rogue capabilities in digitally connected energy devices. Facility infrastructure improvements enhanced the thermal research network, the SCADA system, the cyber range, power hardware-in-the-loop testing, and more. With support from the U.S Department of Energy (DOE), the ESIF laboratories continue to deliver leading solutions for secure, reliable, and affordable power.

24 POWER TRANSMISSION AND DISTRIBUTION↗

DOE OSTI2020

Performance Assessment of AFQMC Implementation (ECP Milestone 5.2 Report)

The auxiliary field quantum Monte Carlo algorithm in QMCPACK was ported to run on AMD GPUs using HIP. Initial performance measurements on early access hardware indicate a 4x slowdown compared to NVIDIA V100s. The slowdown was determined to be a result of suboptimal ROCM libraries for small matrix sizes coupled with some poorly performing hipified CUDA kernels. Through a preliminary analysis of HIP kernel performance we determined a strategy for pro ling and tuning kernels. We request that ROCM libraries are optimized for smaller problem sizes particularly for batched operations.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2012

ATCA/muTCA for Physics

ATCA/{mu}TCA platforms are attractive because of the modern serial link architecture, high availability features and many packaging options. Less-demanding availability applications can be met economically by scaling back speed and redundancy. The ATCA specification was originally targeted for the Telecom industry but has gained recently a much wider user audience. The purpose of this paper is to report on present hardware and software R and D efforts where ATCA and {mu}TCA are planned, already being used or in development using selected examples for accelerator and detectors in the Physics community. It will present also the status of a proposal for physics extensions to ATCA/{mu}TCA specifications to promote inter-operability of laboratory and industry designs for physics.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2021

A review of machine learning in building load prediction

The surge of machine learning in recent years has been empowering engineer modeling in various fields. The decreasing hardware cost, increasing data accessibility, and advances of building automation system (BAS) allow the collection and storage of a significant amount of building operation data. The two facts provide great opportunities of applying machine learning to building energy systems modeling and analysis. There are a great number of research papers on this topic but there lacks a comprehensive and general review to summarize the current development, limitations, gaps and future trend. In this review paper series, machine learning techniques in building energy system modeling and analysis are reviewed under the organization and logic of the machine learning definition by Tom M. Mitchell: a computer program is said to learn from experience E with respect to some class of tasks T and performance measure P if its performance at tasks in T, as measured by P, improves with experience E. This paper is the first part of the review paper series, which focuses on building load prediction. First, the applications of building load prediction model (task T) are reviewed. Then, the modeling algorithms improving machine learning performance and accuracy (performance P) are reviewed. At the same time, the literature on the data perspective for modeling (experience E), including data engineering from sensors level to data level, pre-processing, feature extraction and selection, is reviewed. Finally, what is well-studied and what is lacking but with great potential are concluded; the gaps between present and future utilization of machine learning techniques are identified; the future trend and development are also predicted. The target readers of this paper are not only researchers from the building side who can get exposed to cutting edge machine learning tools, but also those from machine learning side who can understand the potential and challenge to apply machine learning in buildings.

Liang, Zhang↗

DOE OSTI2024

Impacts of floating-point non-associativity on reproducibility for HPC and deep learning applications

Run to run variability in parallel programs caused by floating-point non-associativity has been known to significantly affect reproducibility in iterative algorithms, due to accumulating errors. Non-reproducibility can critically affect the efficiency and effectiveness of correctness testing for stochastic programs. Recently, the sensitivity of deep learning training and inference pipelines to floating-point non-associativity has been found to sometimes be extreme. It can prevent certification for commercial applications, accurate assessment of robustness and sensitivity, and bug detection. New approaches in scientific computing applications have coupled deep learning models with high-performance computing, leading to an aggravation of debugging and testing challenges. Here we perform an investigation of the statistical properties of floating-point non-associativity within modern parallel programming models, and analyze performance and productivity impacts of replacing atomic operations with deterministic alternatives on GPUs. We examine the recently-added deterministic options in PyTorch within the context of GPU deployment for deep learning, uncovering and quantifying the impacts of input parameters triggering run to run variability and reporting on the reliability and completeness of the documentation. Finally, we evaluate the strategy of exploiting automatic determinism that could be provided by deterministic hardware, using the Groq LPUTM accelerator for inference portions of the deep learning pipeline. We demonstrate the benefits that a hardware-based strategy can provide within reproducibility and correctness efforts.

Shanmugavelu, Sanjif↗

DOE OSTI2017

Building Energy Management Open Source Software

This is the repository for Building Energy Management Open Source Software (BEMOSS), which is an open source operating system that is engineered to improve sensing and control of equipment in small- and medium-sized commercial buildings. BEMOSS offers the following key features: (1) Open source, open architecture – BEMOSS is an open source operating system that is built upon VOLTTRON – a distributed agent platform developed by Pacific Northwest National Laboratory (PNNL). BEMOSS was designed to make it easy for hardware manufacturers to seamlessly interface their devices with BEMOSS. Software developers can also contribute to adding additional BEMOSS functionalities and applications. (2) Plug & play – BEMOSS was designed to automatically discover supported load controllers (including smart thermostats, VAV/RTUs, lighting load controllers and plug load controllers) in commercial buildings. (3) Interoperability – BEMOSS was designed to work with load control devices form different manufacturers that operate on different communication technologies and data exchange protocols. (4) Cost effectiveness – Implementation of BEMOSS deemed to be cost-effective as it was built upon a robust open source platform that can operate on a low-cost single-board computer, such as Odroid. This feature could contribute to its rapid deployment in small- or medium-sized commercial buildings. (5) Scalability and ease of deployment – With its multi-node architecture, BEMOSS provides a distributed architecture where load controllers in a multi-floor and high occupancy building could be monitored and controlled by multiple single-board computers hosting BEMOSS. This makes it possible for a building engineer to deploy BEMOSS in one zone of a building, be comfortable with its operation, and later on expand the deployment to the entire building to make it more energy efficient. (6) Ability to provide local and remote monitoring – BEMOSS provides both local and remote monitoring ability with role-based access control. (7) Security – In addition to built-in security features provided by VOLTTRON, BEMOSS provides enhanced security features, including BEMOSS discovery approval process, encrypted core-to-node communication, thermostat anti-tampering feature and many more. (8) Support from the Advisory Committee – BEMOSS was developed in consultation with an advisory committee from the beginning of the project. BEMOSS advisory committee comprises representatives from 22 organizations from government and industry.

Source record↗

DOE OSTI2025

Optimized V1G and V2G Electric Vehicle Fleet Management and Grid Transaction at Marine Corps Air Station Miramar in San Diego, CA

The overall technical goal of the project was to demonstrate an all-electric bi-directional non-tactical fleet at Marine Corps Air Station (MCAS) Miramar that was integrated and controlled with other distributed energy resources (DERs) (i.e., PV, stationary battery, and building loads) to provide resilience to critical electric loads in the event of grid outages, to minimize charging costs, and to provide economic energy resources to electricity markets. In this project, the specific, technical objectives were: 1. Demonstrate that bi-directional electric vehicles can provide critical complementary services to fixed storage batteries in microgrid applications while performing function as non-tactical vehicles. 2. Demonstrate participation of bi-directional (V2G) and unidirectional (V1G) PEVs for demand management and minimization of charging costs. 3. Demonstrate integration of multiple DERs for grid service participation. US Marine Corps Air Station (MCAS) Miramar in San Diego was the site of this electric vehicle-to-microgrid-utility grid test and demonstration project. Existing microgrid assets in this study included (1) a public works building; (2) a 30-kW rooftop photovoltaic (PV) system and (3) a separate 250 kW carport PV system. In this project, six bi-directional V2G vans were located at the MCAS Miramar’s showcase building-scale microgrid to develop and test technical capabilities that V2G can provide in microgrid applications (e.g., cost reduction and resiliency). These resources provided aggregated demand management and simulated participation in current retail DR programs. The vehicles used in this demonstration were selected because they provided functionality that MCAS Miramar needed, 15 passenger transport and facilities work cargo carrying capacity, and bi-directional charging capability that the research project required. All vehicles in this study were manufactured and distributed by VIA Motors, Inc. There were six vehicles total and each was VIA’s VTRUX eREV V2G model, a modified General Motors Chevrolet 2500 2WD van. Three of the vans were configured as passenger vans and the other three were configured as cargo vans. Each van had an on-board bi-direcrtional inverter/charger, Bel Power Solutions model 350INVCHGT150-120-240-8G nominally rated at +/-15 kW. The VIA van’s charging connector follows the J1772 charging protocol. The bi-directional EVSEs demonstrated in this study were manufactured by Coritech, Inc. Each VGI-80-AC charging station enabled enhanced V2G charging capability to a Clipper Creek CS-100 charging module. The enhanced capabilities included ethernet communication following the SEP2.0 protocol with a distributed energy resource function set and an operator screen displaying real-time SOC, voltage, and current. The VGI-80-AC charging stations are classified as level 2 with a maximum current output of 80 A or effectively 19 kW. The VIA van’s onboard charger limited the charging and discharging power to 15 kW in each direction. A control computer was installed in the EWOC and connected to an existing monitor. The V2G control communication network was a completely stand-alone closed system that did not have any connection to any other networks on the base. A cybersecure remote communication connection was created with a cellular modem, firewall hardware, and a virtual private network configuration.

24 POWER TRANSMISSION AND DISTRIBUTION↗

DOE OSTI2017

Building Energy Management Open Source Software

Source record↗

DOE OSTI2025

CRCNS22 Learning Rules in the Hippocampus and their Mapping to Neuromorphic Systems (Final Technical Report)

Large scale biologically-realistic computational models are key to investigating the interplay between structure and function in nervous systems, thus paving the way to new clinical methods and neuro-inspired computing solutions. This project focuses on the hippocampus, in particular the CA3-CA1 regions, due to their role in associative learning and memory, pattern separation and completion, and spatial navigation. Investigations into the neuronal organization and learning rule(s) of this circuit can shed light into how declarative memories are formed, stored, recalled and forgotten and inform computational, experimental and clinical neuroscience work. Our project aims at developing a novel data-driven methodology supported by a broad heterogeneous base of neuroscience experimental knowledge and inspired from advances in computer science and engineering. Specifically, this work will benchmark existing and new learning rules within a full-scale spiking neural network simulation of the CA3-CA1 region. The model will be based on an open-source repository, called the Hippocampome, which contains neuronal morphologies, firing patterns, synapse probabilities, and most other required parameters for all known neuron types in the rodent hippocampal formation. The model will be first trained in a supervised fashion for associative memory tasks using backpropagation through time traditionally used in computer science, enhanced with a new technique called the surrogate gradient method. This optimization method will be used to obtain a global loss minimization, but it is not biologically inspired as it assumes the use of data not locally available to the synapses. However, we propose its use as a benchmarking tool, to compare the training performance of local biologically plausible and hardware-mappable learning rules at scale. New rules or combinations will be proposed and tested as needed, based on the obtained results. Progress in this area will also drive the development of novel hardware-mappable algorithms for continual lifelong learning and categorization of new events from few presented examples. This project goes beyond the existing state-of-the-art by looking at large scale realistic neuronal circuits as networks trainable via global optimization methods such as surrogate gradient descent. The objective function of the brain that supports learning is largely unknown, but it is likely that it operates through local learning rules. Studying network trajectories around local minima as proposed in this work represents a useful strategy for understanding whether a network is training by using a specific (set of) learning rule(s). Starting from a completely untrained network is a challenging test since it is difficult to determine how the learning rule affects the trajectory of the network. This interdisciplinary project will help understand what rule governs learning in these regions or if multiple learning rules are involved. The work will develop a robust methodology to measure if the network is converging to the target solution, oscillating around it, or diverging away.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2014

Model-based feasibility assessment and evaluation of prostate hyperthermia with a commercial MR-guided endorectal HIFU ablation array

Purpose: Feasibility of targeted and volumetric hyperthermia (40–45 °C) delivery to the prostate with a commercial MR-guided endorectal ultrasound phased array system, designed specifically for thermal ablation and approved for ablation trials (ExAblate 2100, Insightec Ltd.), was assessed through computer simulations and tissue-equivalent phantom experiments with the intention of fast clinical translation for targeted hyperthermia in conjunction with radiotherapy and chemotherapy. Methods: The simulations included a 3D finite element method based biothermal model, and acoustic field calculations for the ExAblate ERUS phased array (2.3 MHz, 2.3 × 4.0 cm{sup 2}, ∼1000 channels) using the rectangular radiator method. Array beamforming strategies were investigated to deliver protracted, continuous-wave hyperthermia to focal prostate cancer targets identified from representative patient cases. Constraints on power densities, sonication durations and switching speeds imposed by ExAblate hardware and software were incorporated in the models. Preliminary experiments included beamformed sonications in tissue mimicking phantoms under MR temperature monitoring at 3 T (GE Discovery MR750W). Results: Acoustic intensities considered during simulation were limited to ensure mild hyperthermia (T{sub max} < 45 °C) and fail-safe operation of the ExAblate array (spatial and time averaged acoustic intensity I{sub SATA} < 3.4 W/cm{sup 2}). Tissue volumes with therapeutic temperature levels (T > 41 °C) were estimated. Numerical simulations indicated that T > 41 °C was calculated in 13–23 cm{sup 3} volumes for sonications with planar or diverging beam patterns at 0.9–1.2 W/cm{sup 2}, in 4.5–5.8 cm{sup 3} volumes for simultaneous multipoint focus beam patterns at ∼0.7 W/cm{sup 2}, and in ∼6.0 cm{sup 3} for curvilinear (cylindrical) beam patterns at 0.75 W/cm{sup 2}. Focused heating patterns may be practical for treating focal disease in a single posterior quadrant of the prostate and diffused heating patterns may be useful for heating quadrants, hemigland volumes or even bilateral targets. Treatable volumes may be limited by pubic bone heating. Therapeutic temperatures were estimated for a range of physiological parameters, sonication duty cycles and rectal cooling. Hyperthermia specific phasing patterns were implemented on the ExAblate prostate array and continuous-wave sonications (∼0.88 W/cm{sup 2}, 15 min) were performed in tissue-mimicking material with real-time MR-based temperature imaging (PRFS imaging at 3.0 T). Shapes of heating patterns observed during experiments were consistent with simulations. Conclusions: The ExAblate 2100, designed specifically for thermal ablation, can be controlled for delivering continuous hyperthermia in prostate while working within operational constraints.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

DOE OSTI2025

Obstacles to Practical Digital Supply Chain Risk Management in the Energy Sector

Cyber supply chain risk management (C-SCRM) programs must consider operations that depend on the lifecycles of digital components such as hardware, firmware, software, and services. We integrate academic literature, historical incidents, and existing standards to identify obstacles faced by C-SCRM programs.

Business Process Management & Integration↗

DOE OSTI2022

Resilience Design Patterns: A Structured Approach to Resilience at Extreme Scale (V.2.0)

Reliability is a serious concern for future extreme-scale high-performance computing (HPC) systems. Projections based on the current generation of HPC systems and technology roadmaps suggest the prevalence of very high fault rates in future systems. The errors resulting from these faults will propagate and generate various kinds of failures, which may result in outcomes ranging from result corruptions to catastrophic application crashes. Therefore, the resilience challenge for extreme-scale HPC systems requires coordination between various hardware and software technologies that are capable of handling a broad set of fault models at accelerated fault rates. Also, due to practical limits on power consumption in future HPC systems, they are likely to embrace innovative architectures, increasing the levels of hardware and software complexities. Therefore, the techniques that seek to improve resilience must navigate the complex trade-off space between resilience and the overheads to power consumption and performance. While the HPC community has developed various resilience solutions, application-level techniques as well as system-based solutions, the solution space of HPC resilience techniques remains fragmented. There are no formal methods to integrate the various HPC resilience techniques into composite solutions, nor are there methods to holistically evaluate the adequacy and efficacy of such solutions in terms of their protection coverage, and their performance & power efficiency characteristics. Additionally, few implementations of current resilience solutions are portable to newer architectures and software environments that will be deployed on future systems. We developed a new structured approach to the management of HPC resilience using the concept of resilience-based design patterns. In general, a design pattern is a repeatable solution to a commonly occurring problem. We identified the well-known solutions that are commonly used to deal with faults, errors and failures in HPC systems. In the initial design patterns specification (version 1.0), we described the various solutions, which address specific problems in the design of resilient HPC environments, in the form of patterns. Each pattern describes a problem caused by a fault, error or failure event in an HPC environment, and then describes the core of the solution of the problem in such a way that this solution may be adapted to different systems and implemented at different layers of the system stack. The catalog of these resilience design patterns provides designers with a collection of design elements. To construct complete resilience solutions using combinations of various patterns, we defined a framework that enhances HPC designers' understanding of the important constraints and the opportunities for the design patterns to be implemented and deployed at various layers of the system stack. The design framework is also useful for establishing interfaces and mechanisms to coordinate flexible fault management across hardware and software components, as well as to consider the trade-off between performance, resilience, and power consumption when constructing a solution. The resilience design patterns specification version 1.1 included more detailed explanations of the pattern solutions, the context in which the patterns are applicable, and the implications for hardware or software design. It also provided several additional examples and detailed case studies to demonstrate the use of patterns to build realistic solutions. In version 1.2 of the specification document, we have improved the pattern descriptions, including graphical representations of the pattern components. These improvements are largely based on critical comments, feedback and suggestions received from pattern experts and readers of the previous versions of the specification. The pattern classification has been modified to further clarify the relationships between pattern categories. This version of the specification also introduces a pattern language for resilience design patterns. The pattern language presents the patterns in the catalog as a network, revealing the relations among the resilience patterns. The language provides designers with the means to explore alternative techniques for handling a specific fault model that may have different efficiency and complexity characteristics. Using the pattern language also enables the design and implementation of comprehensive resilience solutions as a set of interconnected resilience patterns that can be instantiated across layers of the system stack. The overall goal of this work is to provide hardware and software designers, as well as the users and operators of HPC systems, a systematic methodology for the design and evaluation of resilience technologies in HPC systems that keep scientific applications running to a correct solution in a timely and cost-efficient manner despite frequent faults, errors, and failures of various types. Version 2.0 expands the resilience design pattern classification and catalog to include self-stabilization patterns and reliability, availability and performance models for each structural pattern.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2024

Development of a Hardware-in-The-Loop Testbed for a Decentralized, Data-Driven Electric Vehicle Charging Control Algorithm

This study presents the design of an electric vehicle (EV)-grid integration (EVGI) hardware test-bed to implement smart EV charging algorithms. Here, the proposed test-bed also allows to create different grid events via flexible integration of other power hardware (e.g., controllable loads and battery energy storage systems) and test their impacts on EV charging. The design uses a real-time digital simulator to realize a complex distribution grid model with primary and secondary networks. A grid simulator physically realizes the selected nodes of the simulated grid to power an actual EV, forming a hardware-in-the-loop (HIL) test setup. The EV-grid integration is demonstrated based on the custom hardware and software implementation of the J1772 charging protocol using dSPACE MicroLabBox, operating as a custom EV Supply Equipment (EVSE). The HIL test-bed features a novel testing platform for accurate implementation and analysis of scalable charging algorithms. To this end, a data-driven, decentralized, model-free charging controller based on the Additive Increase and Multiplicative Decrease (AIMD) algorithm is presented and validated on an EV using the HIL test-bed. We tested the proposed algorithm under various case studies, and presented a comparison study with an existing droop-based, decentralized charging solution. The results showed that the EV successfully performed charging commands generated by the EVSE and regulated its charging power to effectively reduce the system loading caused by high EV penetration.

33 ADVANCED PROPULSION SYSTEMS↗

DOE OSTI2023

Architecture for Quantum-in-the Loop Real-Time Simulations for Designing Resilient Smart Grids

With the power grid growing more complex every day with the inclusion of new sensors and regulatory approvals that enable end-users and small local developers to participate in the grid, it is becoming challenging for conventional smart grid simulation, emulation, and testing technologies to keep up. In this work, we propose that quantum-encoded real-time simulations can be helpful under the new paradigm and operational circumstances to solve optimization problems for power grids. By leveraging the principles of quantum mechanics, the proposed quantum-in-loop (QIL) framework will enable better and faster optimization solutions based on real-world, real-time data streams, facilitating the real-time planning and operations of electrical grids that rely on millions of distributed sensors and controllers. Furthermore, QIL will allow researchers and engineers to assist utilities in designing, developing, and de-risking algorithms to optimize power grid operation and resilience by considering inputs from millions of grid-connected devices. QIL framework is being developed to have a self-limiting triage mechanism, which will help engineers and practitioners identify fundamental physical limits on quantum processors, revealing what quantum algorithms can and cannot do in utility-specific use cases and must continue to count on classical high-performance computing infrastructure.

digital real-time simulation↗

DOE OSTI2022

Charge-Based Droop Control Addressing Control Saturation for Low-Inertia Converters

With bulky dc-link capacitors/inductors, traditional voltage-source converters (VSCs) and current-source converters (CSCs) feature a large inertia to facilitate the converter control under large transients. To achieve high power density with reduced cost, low-inertia converters (LICs) featuring significantly reduced dc-link capacitors/inductors have attracted growing attention. However, without bulky dc-link energy buffer, LICs are prone to control saturation under large transients, resulting in undesired oscillation and instability. This issue cannot be managed by traditional proportional-integral based control due to the low inertia. And it will deteriorate when several LICs are connected in series. To address this challenge in LICs, a model-predictive control (MPC) with a computation-inexpensive feed-forward compensation method has been proposed to provide fast dynamic responses. But it would suffer control saturation under larger transients that degrades the control performance. In this article, a charge-based droop control (CDC) is proposed to address this remaining challenge. This paper firstly analyzes the control saturation challenge in LICs by using a tri-port soft-switching solid-state transformer as an example. Next, the operating principle of the proposed CDC are introduced. Two different implementation approaches are discussed in detail. Lastly, the proposed scheme is validated in simulation with a high-fidelity model of hardware prototype. As a result, the proposed CDC eliminated the 2 kHz oscillation and reduced the dc-link ripple and overshoot due to control saturation by 75% and 50%, respectively.

30 DIRECT ENERGY CONVERSION↗

DOE OSTI2018

Data Analyst - A Device for Continuous Repeating Real-Time Assay of HPGe, CZT, or Scintillation Gamma Spectra - 18574

There are many situations where a continuous repeating series of quantitative gamma spectral assays is desirable. These include measurements of fluid going through a pipe, of items on a conveyor passing under a detector, a detector aimed at the ground on a moving platform, or a detector in free air as a plume of radioactivity drifts by. While Canberra and many others have built and deployed many of these devices, they have all been custom built for that particular application, and rather expensive. The Data Analyst [DA] is a general purpose device that is designed to fulfill many of these applications without modification, but can be fairly-easily modified if needed. The DA is a small [12 x 18 x 5 cm], low-power [3 W] box that accepts the input from external MCAs, controls the acquisition cycle, analyzes the results, and transmits them to the outside world when needed. The detector-MCAs supported are HPGe with Lynx MCA, NaI or CeBr or LaBr with Osprey MCA, and CZT or CsI with internal MCAs. The DA internal software used for analysis is the full standard Genie analysis software, which is very widely used and well accepted in the industry. This allows exported spectral files to be reviewed and confirmed by Experts, and reanalyzed if needed. The DA hardware is totally autonomous. As soon as power is applied, it starts the pre-programmed sequences of acquisition and analysis. Multiple analysis sequences [called Workflows] can be in operation simultaneously, e.g. with different count times and different nuclide libraries. Only the detector with MCA, and the DA are required in the field. A PC can be connected to the DA locally or remotely via WiFi, USB, or Ethernet. The PC is needed to setup the DA, and to readout the results. The DA also supports [with application-specific programming] external Inputs/Outputs either for recording with the analysis results or for use in the computation of the result, e.g. temperature, flow rate, pressure, ... ]. The DA also has an internal GPS receiver with external antenna. Applications of the DA that are discussed include stack gas effluent activity where a portion of the effluent stream was extracted to flow through a shield with a HPGe detector and Lynx MCA, and measurement of reactor coolant activity where a shielded CZT detector and MCA were aimed at various sections of reactor primary coolant during a maintenance outage. Potential other applications are also described. (author)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

DOE OSTI2021

Intrinsic synaptic plasticity of ferroelectric field effect transistors for online learning

Nanoelectronic devices emulating neuro-synaptic functionalities through their intrinsic physics at low operating energies is imperative toward the realization of brain-like neuromorphic computers. In this work, we leverage the non-linear voltage dependent partial polarization switching of a ferroelectric field effect transistor to mimic plasticity characteristics of biological synapses. Here, we provide experimental measurements of the synaptic characteristics for a 28nm high-k metal gate technology based device and develop an experimentally calibrated device model for large-scale system performance prediction. Decoupled read-write paths, ultra-low programming energies and the possibility of arranging such devices in a cross-point architecture demonstrate the synaptic efficacy of the device. Our hardware-algorithm co-design analysis reveals that the intrinsic plasticity of the ferroelectric devices has potential to enable unsupervised local learning in edge devices with limited training data

42 ENGINEERING↗

DOE OSTI2018

Performance study of atomic tally methods for GPU-accelerated Monte Carlo dose calculation - 25352

Over the past several years, the graphics processing unit (GPU) technology has rapidly gained ground in scientific computing due to its outstanding performance and programmability. GPU implementation of Monte Carlo radiation transport for dose calculations has been reported by many investigators. The majority of these studies adopted single-precision floating point format because of the higher peak floating point operations per second (FLOPS) the GPUs can deliver than double-precision. It has been known that calculation using single-precision is more prone to numerical round-off errors, especially when a single tally data is accumulated 'atomically' and repeatedly by thousands of GPU threads. To mitigate this problem, the least intrusive solution in theory is to replace the single-precision atomic-add tally function with a double-precision version. However, the complexity lies in the fact that some GPUs (Nvidia GPUs prior to the Pascal generation; all current AMD GPUs) do not readily offer such double-precision function at hardware level, and that software emulation is too slow to use if not optimized properly. This paper discusses several atomic-add tally methods with reduced numerical errors used throughout ARCHER development. The original software-based compare-and-swap method (CAS) was shown to be inefficient due to high intra-warp thread contention, whereas the improved software-based warp-aggregated method (WAG) and Kahan summation method (KAS) eliminated the thread contention and performed very well on Kepler and Maxwell GPUs, being more than 13 times faster than CAS in our tests. The hardware-based (HB) double-precision atomic-add feature available on Pascal and Volta GPUs exceeded software emulation and offered the best performance universally, so did WIB, a combination of WAG and HB methods. It was also shown that in a single-precision dose engine, KAS managed to maintain high reliability when the single-precision atomic-add tally underestimated the dose by 25%. (authors)

61 RADIATION PROTECTION AND DOSIMETRY↗