SEARCH · Engineering Papers

Results for “network acceleration”

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 757 records · Page 42

All sources NASA NTRS DOE OSTI

DOE OSTI2025

Defect modeling in semiconductors: the role of first principles simulations and machine learning

Abstract Point defects in semiconductors dictate their electronic and optical properties. Vacancies, interstitials, substitutional defects, and defect complexes can form in the semiconductor lattice and significantly impact its performance in applications such as solar absorption, light emission, electronics, and catalysis. Understanding the nature and energetics of point defects is essential for the design and optimization of next-generation semiconductor technologies. Here, we provide a comprehensive overview of the current state of research on point defects in semiconductors, focusing on the application of density functional theory (DFT) and machine learning (ML) in accelerating the prediction and understanding of defect properties. DFT has been instrumental in accurately calculating defect formation energies, charge transition levels, and other defect-related properties such as carrier recombination rates and lifetimes, and ion migration barriers. ML techniques, particularly neural networks, have emerged as powerful tools for enabling rapid prediction of defect properties at DFT-accuracy in order to overcome the expense of using large supercells and advanced functionals. We begin this article with a discussion of different types of point defects and complexes, their impact on semiconductor properties, and the experimental and DFT approaches typically used for their characterization. Through multiple case studies, we explore how DFT has been successfully applied to understand defect behavior across a variety of semiconductors, and how ML approaches integrated with DFT can efficiently predict defect properties and facilitate the discovery of new materials with tailored defect behavior. Overall, the advent of ‘DFT+ML’ promises to drive advancements in semiconductor technology, catalysis, and renewable energy applications, paving the way for the development of high-performance semiconductors which are defect-tolerant or have desirable dopability.

Rahman, Md Habibur (ORCID:000000027705984X)↗

DOE OSTI2024

Low Power, Radiation Resilient Synchronous Edge Processing for Remote Monitoring

Next-generation space remote sensing systems may be equipped with imaging arrays that sense data at a rate that outstrips the processing capability of any computing hardware that can operate within a satellite’s power budget. This project developed novel convolutional and recurrent neural networks to detect and estimate point-like events amid clutter, and investigated their efficient and accurate implementation on analog in-memory computing systems that are 10-1000× more energy-efficient than digital processors. This project leveraged two memory devices at different levels of technological maturity: a large-scale analog computing prototype using commercial SONOS charge-trap memory, and electrochemical memory (ECRAM) with intrinsic radiation hardness. We experimentally demonstrated end-to-end analog processing of our neural networks on SONOS and characterized the radiation response of both SONOS and ECRAM. We advanced the state-of-the-art in ECRAM precision and reliability, and developed co-design methods to enable accurate long-term operation of SONOS analog accelerators in space radiation environments.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2022

Seismic Waveform Inversion Capability on Resource-Constrained Edge Devices

Seismic full wave inversion (FWI) is a widely used non-linear seismic imaging method used to reconstruct subsurface velocity images, however it is time consuming, has high computational cost and depend heavily on human interaction. Recently, deep learning has accelerated it’s use in several data-driven techniques, however most deep learning techniques suffer from overfitting and stability issues. In this work, we propose an edge computing-based data-driven inversion technique based on supervised deep convolutional neural network to accurately reconstruct the subsurface velocities. Deep learning based data-driven technique depends mostly on bulk data training. In this work, we train our deep convolutional neural network (DCN) (UNet and InversionNet) on the raw seismic data and their corresponding velocity models during the training phase to learn the non-linear mapping between the seismic data and velocity models. The trained network is then used to estimate the velocity models from new input seismic data during the prediction phase. The prediction phase is performed on a resource-constrained edge device such as Raspberry Pi. Raspberry Pi provides real-time and on-device computational power to execute the inference process. In addition, we demonstrate robustness of our models to perform inversion in the presence on noise by performing both noise-aware and no-noise training and feeding the resulting trained models with noise at different signal-to-noise (SNR) ratio values. We make great efforts to achieve very feasible inference times on the Raspberry Pi for both models. Specifically, the inference times per prediction for UNet and InversionNet models on Raspberry Pi were 22 and 4 s respectively whilst inference times for both models on the GPU were 2 and 18 s which are very comparable. Finally, we have designed a user-friendly interactive graphical user interface (GUI) to automate the model execution and inversion process on the Raspberry Pi.

Manu, Daniel (ORCID:0000000154982677)↗

DOE OSTI2025

dCache: The Storage System of Choice for Data-Intensive Applications

The ever-increasing volumes of data produced by modern scientific facilities like EuXFEL and LHC put significant stress on data management infrastructure operated by laboratories and research centers. The challenges to be addressed span the entire data life cycle, from ingest and efficient data analysis to long-term preservation, typically involving large tape libraries. dCache, a storage system developed in collaboration between the Deutsches Elektronen-Synchrotron (DESY), Fermi National Accelerator Laboratory, and Nordic e-Infrastructure Collaboration (NeIC), is designed to manage a large number of disk servers and to facilitate transparent data migration to and from archival storage. Its multifaceted approach offers a unified method to support a variety of scientific use cases with the same storage infrastructure, including high-throughput data ingest, data sharing over wide area networks, efficient access from HPC clusters, and long-term data preservation on tertiary storage. Initially developed for high energy physics (HEP) experiments, dCache is now used by various scientific communities, including astrophysics, biomedical research, and life sciences, each having specific requirements. This paper presents architecture, deployment strategies, performance and scalability enhancements, and recent advancements in dCache addressing the needs of scientific communities. Finally, we touch on the development and release process, ensuring the software’s high quality.

DCache↗

DOE OSTI2025

Neural refinement of sample weights

Monte Carlo simulations are an essential tool in particle physics data analysis. Events are typically generated alongside weights that redistribute the cross section of the simulated process across the phase space. These weights can be negative, and several post hoc methods have been developed to eliminate or mitigate the negative values. All of these methods share the common strategy of approximating the average weight as a function of phase space. We introduce an alternative approach, which, instead of reweighting to the average, refines the initial weights with a scaling transformation, utilizing a phase space-dependent factor. Since this new refinement method does not need to model the full weight distribution, it can be more accurate. High-dimensional and unbinned phase space is processed using neural networks for the refinement method. In addition to the refinement method, we introduce a new resampling protocol, which can be used in conjunction with any weight transformation to not only preserve the average weight but also the statistical uncertainties of the initial distribution. Using both realistic and synthetic examples, we show that the new neural refinement method is able to match or exceed the accuracy of similar weight transformations and that the new resampling protocol is simpler in implementation than previous methods while exhibiting equivalent statistical properties.

Artificial neural networks↗

DOE OSTI2021

Distributed deep reinforcement learning for simulation control

Abstract Several applications in the scientific simulation of physical systems can be formulated as control/optimization problems. The computational models for such systems generally contain hyperparameters, which control solution fidelity and computational expense. The tuning of these parameters is non-trivial and the general approach is to manually ‘spot-check’ for good combinations. This is because optimal hyperparameter configuration search becomes intractable when the parameter space is large and when they may vary dynamically. To address this issue, we present a framework based on deep reinforcement learning (RL) to train a deep neural network agent that controls a model solve by varying parameters dynamically. First, we validate our RL framework for the problem of controlling chaos in chaotic systems by dynamically changing the parameters of the system. Subsequently, we illustrate the capabilities of our framework for accelerating the convergence of a steady-state computational fluid dynamics solver by automatically adjusting the relaxation factors of the discretized Navier–Stokes equations during run-time. The results indicate that the run-time control of the relaxation factors by the learned policy leads to a significant reduction in the number of iterations for convergence compared to the random selection of the relaxation factors. Our results point to potential benefits for learning adaptive hyperparameter learning strategies across different geometries and boundary conditions with implications for reduced computational campaign expenses 4 4 Data and codes available at https://github.com/Romit-Maulik/PAR-RL . .

42 ENGINEERING↗

DOE OSTI2025

The Upgrade of the SNS Extraction Kicker Power Supplies for the Proton Power Upgrade

The Proton Power Upgrade (PPU) project at the Spallation Neutron Source (SNS) aims to enhance the linear accelerator's power output from 1.4 to 2.8 megawatts. A critical focus lies on upgrading the Extraction Kicker Power Supplies to handle the amplified beam power. The existing capacitor charging power supplies lacked the capability to charge the Blumlein-type Pulse Forming Network (PFN) to the required 40 kV at a repetition rate of 60 Hz. To address this limitation, an initial solution involving the use of a resonant charging circuit was proposed. However, a challenge arose with a self-firing issue in the thyristor switch, which proved unsolvable within the given timeframe. Facing a critical decision point in the project, an alternative approach was explored: integrating a supplementary power supply to enhance the overall power capacity. This solution was tested and ultimately chosen as the path forward. To accommodate the additional power supply, the power distribution system had to undergo upgrades. Throughout this endeavor, a comprehensive examination of the system took place, with meticulous identification of key components. This paper delves into the technical challenges encountered during the process and the corresponding solutions that were implemented.

Tan, Yugang [ORNL] (ORCID:0000000155968252)↗

DOE OSTI2020

Optimizing High Performance Markov Clustering for Pre-Exascale Architectures

HipMCL is a high-performance distributed memory implementation of the popular Markov Cluster Algorithm (MCL) and can cluster large-scale networks within hours using a few thousand CPU-equipped nodes. It relies on sparse matrix computations and heavily makes use of the sparse matrix-sparse matrix multiplication kernel (SpGEMM). The existing parallel algorithms in HipMCL are not scalable to Exascale architectures, both due to their communication costs dominating the runtime at large concurrencies and also due to their inability to take advantage of accelerators that are increasingly popular. In this work, we systematically remove scalability and performance bottlenecks of HipMCL. We enable GPUs by performing the expensive expansion phase of the MCL algorithm on GPU. Additionally, we propose a CPU-GPU joint distributed SpGEMM algorithm called pipelined Sparse SUMMA and integrate a probabilistic memory requirement estimator that is fast and accurate. Furthermore, we develop a new merging algorithm for the incremental processing of partial results produced by the GPUs, which improves the overlap efficiency and the peak memory usage. We also integrate a recent and faster algorithm for performing SpGEMM on CPUs. We validate our new algorithms and optimizations with extensive evaluations. With the enabling of the GPUs and integration of new algorithms, HipMCL is up to 12.4x faster, being able to cluster a network with 70 million proteins and 68 billion connections just under 15 minutes using 1024 nodes of ORNL's Summit supercomputer.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2024

Invertible Temper Modeling using Normalizing Flows and the Effects of Structure Preserving Loss

Advanced manufacturing research and development is typically small-scale, owing to costly experiments associated with these novel processes. Deep learning techniques could help accelerate this development cycle but frequently struggle in small-data regimes like the advanced manufacturing space. While prior work has applied deep learning to modeling visually plausible advanced manufacturing microstructures, little work has been done on data-driven modeling of how microstructures are affected by heat treatment, or assessing the degree to which synthetic microstructures are able to support existing workflows. We propose to address this gap by using invertible neural networks (normalizing flows) to model the effects of heat treatment, e.g., tempering. The model is developed using scanning electron microscope imagery from samples produced using shear-assisted processing and extrusion (ShAPE) manufacturing. This approach not only produces visually and topologically plausible samples, but also captures information related to a sample’s material properties or experimental process parameters. We also demonstrate that topological data analysis, used in prior work to characterize microstructures, can also be used to stabilize model training, preserve structure, and improve downstream results. We assess directions for future work and identify our approach as an important step towards end-to-end deep learning system for accelerating advanced manufacturing research and development.

Howland, Sylvia↗

DOE OSTI2025

Artificial intelligence time series forecasting for feed-forward laser stabilization

Laser plasma accelerators, typically operating at 1–10 Hz repetition rates, have the ability to produce high-quality electron beams in compact, all-optical-driven configurations, with the electron beams uniquely suited for a wide variety of accelerator-based applications. However, fluctuations and drifts in the laser delivery to the meter-scaled and below plasma target (the electron beam source) will translate into electron beam source variations that can limit their utility for demanding applications like light sources or linear colliders. Commercially available active feedback laser stabilization systems are intrinsically bandwidth limited due to their integration with multi-inch corrective mirror mounts which minimizes their effectiveness. In this manuscript, we present a Neural Network time series forecaster that can predict laser position fluctuations of the laser delivery to the final target well ahead of a future laser shot. The Root-Mean-Square-Error (RMSE) of the prediction accuracy was < 2 μ m for a 1 / e 2 beam radius of 34 μ m . Our feed-forward approach serves as a first-step in circumventing the bandwidth limitations imposed by the currently available stabilization systems since it allows for mirrors to be moved into position ahead of time to offset the predicted future position drift. This will help advance laser plasma accelerator research by providing greater robustness and stability needed for its applications.

Berger, Curtis↗

DOE OSTI2025

Electronic structure prediction of medium and high entropy alloys across composition space

We propose machine learning (ML) models to predict the electron density — the fundamental unknown of a material’s ground state — across the composition space of concentrated alloys. From this, other physical properties can be inferred, enabling accelerated exploration. A significant challenge is that the number of descriptors and sampled compositions required for accurate prediction grows rapidly with species. To address this, we employ Bayesian Active Learning (AL), which minimizes training data requirements by leveraging uncertainty quantification capabilities of Bayesian Neural Networks. Compared to the strategic tessellation of the composition space, Bayesian-AL reduces the number of training data points by a factor of 2.5 for ternary (SiGeSn) and 1.7 for quaternary (CrFeCoNi) systems. We also introduce easy-to-optimize, body-attached-frame descriptors, which respect physical symmetries while keeping descriptor-vector size nearly constant as alloy complexity increases. Our ML models demonstrate high accuracy and generalizability in predicting both electron density and energy across composition space.

materials science↗

DOE OSTI2024

Designing workflows for materials characterization

Experimental science is enabled by the combination of synthesis, imaging, and functional characterization organized into evolving discovery loop. Synthesis of new material is typically followed by a set of characterization steps aiming to provide feedback for optimization or discover fundamental mechanisms. However, the sequence of synthesis and characterization methods and their interpretation, or research workflow, has traditionally been driven by human intuition and is highly domain specific. Here, we explore concepts of scientific workflows that emerge at the interface between theory, characterization, and imaging. In this study, we discuss the criteria by which these workflows can be constructed for special cases of multiresolution structural imaging and functional characterization, as a part of more general material synthesis workflows. Some considerations for theory–experiment workflows are provided. We further pose that the emergence of user facilities and cloud labs disrupts the classical progression from ideation, orchestration, and execution stages of workflow development. To accelerate this transition, we propose the framework for workflow design, including universal hyperlanguages describing laboratory operation, ontological domain matching, reward functions and their integration between domains, and policy development for workflow optimization. These tools will enable knowledge-based workflow optimization; enable lateral instrumental networks, sequential and parallel orchestration of characterization between dissimilar facilities; and empower distributed research.

36 MATERIALS SCIENCE↗

DOE OSTI2023

Towards On-Chip Learning for Low Latency Reasoning with End-to-End Synthesis

The Software Defined Architectures (SODA) Synthesizer is an open-source compiler-based tool able to automatically generate domain-specialized systems targeting Application-Specific Integrated Circuits (ASICs) or Field Programmable Gate Arrays (FPGAs) starting from high-level programming. SODA is composed of a frontend, SODA-OPT, which leverages the multilevel intermediate representation (MLIR) framework to interface with productive programming tools (e.g., machine learning frame-works), identify kernels suitable for acceleration, and perform high-level optimizations, and of a state-of-the-art high-level synthesis backend, Bambu from the PandA framework, to generate custom accelerators. One specific application of the SODA Synthesizer is the generation of accelerators to enable ultra-low latency inference and control on autonomous systems for scientific discovery (e.g., electron microscopes, sensors in particle accelerators, etc.). This paper provides an overview of the flow in the context of the generation of accelerators for edge processing to be integrated in transmission electron microscopy (TEM) devices, focusing on use cases from precision material synthesis. We show the tool in action with an example of design space exploration for inference on reconfigurable devices with a conventional deep neural network model (LeNet). Finally, we discuss the research directions and opportunities enabled by SODA in the area of autonomous control for scientific experimental workflows.

Castellana, Vito G.↗

NASA NTRS1992

Battery cars on superconducting magnetically levitated carriers: One commuting solution

Commuting to work in an urban-suburban metropolitan environment is becoming an unpleasant time-wasting process. We applied the technology of communication management to this commuting problem. Communication management is a system-engineering tool that produced today's efficient telephone network. The resulting best commuting option is magnetically levitated carriers of two-passenger, battery-powered, personally-owned local-travel cars. A commuter drives a car to a nearby station, selects a destination, drives on a waiting carrier, and enters an accelerating ramp. A central computer selects an optimum 100 miles-per-hour trunk route, considering existing and forecast traffic; assigns the commuter a travel slot, and subsequently orders switching-station actions. The commuter uses the expensive facilities for only a few minutes during each trip. The cost of travel could be less than 6 cents per mile.

Briggs, B. Mike↗

DOE OSTI2020

Machine Learning for Automated Extraction of Building Geometry

As data science comes to buildings, the promise of using machine learning and novel sources of data has received much attention. Advances in machine learning and computer vision algorithms, combined with increased access to unstructured data (e.g., images and text), have created an opportunity for automated extraction of building characteristics – cost-effectively, and at scale. Acquisition of features such as footprint are time consuming and costly to acquire with today’s manual methods, but can be streamlined through intelligent software-based solutions applied to satellite images. When combined with aerial RGB and thermal images, full 3D geometries and thermal maps can be constructed to determine additional characteristics such as window to wall ratio, height, number of stories and envelope thermal characteristics. In this paper we present three contributions to accelerate these high potential opportunities: (1) a methodical analysis of how these features can be integrated into today’s simulation and data driven software tools to enhance efficiency measure identification and owner/operator decision making; (2) development and accuracy testing of open source deep neural network methods to extract building footprints from satellite imagery, including the curation and application of openly available GIS datasets for training and continued development by others; and (3) an open framework for drone-based image capture and creation of 3D building geometries. This work represents an important bridge between high-level studies that span diverse application areas and those that detail point solutions yet cannot be easily replicated or extended.

Touzani, Samir↗

DOE OSTI2020

Indicator-directed Dynamic Power Management for Iterative Workloads on GPU-Accelerated Systems

Modern high-performance and warehouse computing centers show strong interest in minimizing system power consumption while satisfying customers’ quality of service (QoS). Dynamic voltage and frequency scaling (DVFS) is effective for achieving this goal. Nevertheless, automating the process online and making it transparent to users must address three major challenges: (1) Complexity — today’s hardware components (e.g., CPUs, GPUs, memory, network, etc.) can be configured in several or dozens of frequency/voltage states for satisfying divergent system demands. Given their combination and the emergence of heterogeneity, searching the optimal configuration in the design space online can be timing consuming. (2) QoS guarantee — user-defined objectives such as power constraint and performance target must be monitored, predicted and ensured at the best effort. (3) Adaptability — various known and unknown workloads run on systems. Workloads characteristics should be quickly determined and configurations dynamically adjusted in accord with workloads and QoS. In this work, we focus on applications exhibiting an interesting feature – iterative or periodic, which is common among conventional HPC and emerging machine learning workloads. We propose an online dynamic power-performance (ODPP) management framework to dynamically adjust GPU DVFS configurations to meet performance and power objectives and constraints, without any code annotation or intrusion. Particularly, ODPP extracts the performance and power indicators for applications from their resources utilization profiles in a short episode. It further automatically constructs an accurate model that infers from the indicators how the application's performance and power vary with GPU core and memory frequencies. Aided with the model, for both seen and unseen applications, ODPP can quickly determine the most appropriate DVFS configuration for their execution. We evaluate ODPP on an NVIDIA GPU using multiple exascale computing (ECP) and deep learning applications.

Zou, Pengfei↗

DOE OSTI2023

Early Fault Detection in Particle Accelerator Power Electronics Using Ensemble Learning

Early fault detection and fault prognosis are crucial to ensure efficient and safe operations of complex engineering systems such as the Spallation Neutron Source (SNS) and its power electronics (high voltage converter modulators). Following an advanced experimental facility setup that mimics SNS operating conditions, the authors successfully conducted 21 early fault detection experiments, where fault precursors are introduced in the system to a degree enough to cause degradation in the waveform signals, but not enough to reach a real fault. Nine different machine learning techniques based on ensemble trees, convolutional neural networks, support vector machines, and hierarchical voting ensembles are proposed to detect the fault precursors. Although all 9 models have shown a perfect and identical performance during the training and testing phase, the performance of most models has decreased in the next test phase once they got exposed to realworld data from the 21 experiments. The hierarchical voting ensemble, which features multiple layers of diverse models, maintains a distinguished performance in early detection of the fault precursors with 95% success rate (20/21 tests), followed by adaboost and extremely randomized trees with 52% and 48% success rates, respectively. The support vector machine models were the worst with only 24% success rate (5/21 tests). The study concluded that a successful implementation of machine learning in the SNS or particle accelerator power systems would require a major upgrade in the controller and the data acquisition system to facilitate streaming and handling big data for the machine learning models. In addition, this study shows that the best performing models were diverse and based on the ensemble concept to reduce the bias and hyperparameter sensitivity of individual models.

43 PARTICLE ACCELERATORS↗

DOE OSTI2021

Accelerating phase-field-based microstructure evolution predictions via surrogate models trained by machine learning methods

Abstract The phase-field method is a powerful and versatile computational approach for modeling the evolution of microstructures and associated properties for a wide variety of physical, chemical, and biological systems. However, existing high-fidelity phase-field models are inherently computationally expensive, requiring high-performance computing resources and sophisticated numerical integration schemes to achieve a useful degree of accuracy. In this paper, we present a computationally inexpensive, accurate, data-driven surrogate model that directly learns the microstructural evolution of targeted systems by combining phase-field and history-dependent machine-learning techniques. We integrate a statistically representative, low-dimensional description of the microstructure, obtained directly from phase-field simulations, with either a time-series multivariate adaptive regression splines autoregressive algorithm or a long short-term memory neural network. The neural-network-trained surrogate model shows the best performance and accurately predicts the nonlinear microstructure evolution of a two-phase mixture during spinodal decomposition in seconds, without the need for “on-the-fly” solutions of the phase-field equations of motion. We also show that the predictions from our machine-learned surrogate model can be fed directly as an input into a classical high-fidelity phase-field model in order to accelerate the high-fidelity phase-field simulations by leaping in time. Such machine-learned phase-field framework opens a promising path forward to use accelerated phase-field simulations for discovering, understanding, and predicting processing–microstructure–performance relationships.

36 MATERIALS SCIENCE↗