SEARCH · Search NASA

Results for “sensor data validation”

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 379 records · Page 21

All sources NASA NTRS DOE OSTI

DOE OSTI2021

Sensor enabled data-driven predictive analytics for modeling and control with high penetration of DERs in distribution systems

The electric power grid is undergoing a tremendous transformation due to the increasing penetration of renewable energy resources beginning with wind and more recently with the distributed energy resources (DERs) such as solar and battery storage. DERs have dramatically changed the role of the distribution systems in the overall power grid, and they are expected to contribute a significant portion of power generation in the future. If current trends for DERs continue, system operation and control will need to change dramatically for improved grid reliability and resiliency. As renewable resources increase in penetration, new and challenging operational, planning, and design problems are expected to emerge. Some of the key challenges that arise in the planning and operation of the future grid are: 1) Quantifying the impact of high DER penetration in distribution systems on bulk grid behavior over multiple time scales. 2) Identifying whether a particular DER configuration/settings have a large impact on the overall grid behavior. These challenges can be addressed in an offline manner using detailed T&D grid models and they can also be addressed in an online manner using sensor measurements. In particular, the advancement and planned growth in sensor technology in power grid over various voltage levels provide us with a unique opportunity to tackle these challenges from a data analytic perspective without needing detailed T&D grid models. A few questions that naturally arise when addressing the challenges from DERs using sensor data are: 1) How can we use limited sensor measurements to monitor & control voltage stability and small signal stability of the bulk system? 2) How can we ensure that the developed data analytic methods are robust to data availability and quality issues? 3) How can we compute the developed analytics in a scalable manner using streaming measurements? In this project, we addressed the aforementioned challenges arising from DERs and answered the questions raised above on how to effectively use the sensor measurements to enhance the reliability and performance of the electric grid. Thus, the overarching goal of this project is to develop effective reduced/representative system models from data that make the computational complexity sufficiently manageable so as to be useful to simulate, analyze, and even control complex non-linear power systems dynamics with large penetrations of DERs. In order to achieve the objective, the project team established a four-fold technical approach 1) Formulated a combined transmission-distribution co-simulation framework for data generation and validation, 2) Derived reduced/representative models of power systems based on data-driven methods for efficient computation and appropriate representation of system behavior, 3) Developed data driven characterization of power system behavior based on transfer operator theory, machine learning and optimization for model estimation, 4) Incorporated a scalable data management and processing architecture using distributed Kafka streaming applications that coordinate input data streams to the developed data analytics. The key accomplishments of the project are: 1) Development of a scalable multi-timescale T&D co-simulation framework (both for steady state and for dynamic co-simulation) using commercial solvers (PSSE and GridLAB-D). The steady-state T&D co-simulation interface is shared with our industry partner (PJM). 2) A structured reduced order dynamic model of distribution systems that can represent partial motor stalling along with a systematic procedure to derive the model parameters. 3) A PMU based online method to monitor, localize and mitigate fault-induced delayed voltage recovery using DER reactive support and load control in distribution systems. 4) Development of linear operator based robust methodologies for dynamic state estimation, uncertainty quantification, system identification and trajectory prediction for power system dynamics. 5) An adaptive damping control for utilizing wind energy resources to provide oscillation damping and system stability. 6) Implementation of Kafka-based framework for efficient processing of streaming data using Linux-based local virtual environment.

DER integration↗

DOE OSTI2019

Capacitance-Based Dimensional Change Sensors for In-Pile Materials Measurements

Advanced fuel compositions, such as accident tolerate fuels (ATF), are an active area of developing in the nuclear power industry. The long-term performance of these newly developed fuels is estimated through physics-based simulation models of irradiation-, temperature-, pressure-, etc.-induced material degradation. As these fuels are deployed in test reactors, measurement and characterization of the fuel pin evolution is used to validate prediction models. In-pile material evolution parameters, such as fuel rod pressurization, fuel stack and cladding elongation, and cladding diameter, are commonly measured using a linear voltage differential transformer (LVDT). However, LVDTs are bulky and limited to lower (350-500C) temperature operation. The high power density and small size of most experimental positions in high performance research reactors used for accelerated materials irradiation studies generally precludes the use of LVDTs in these reactors. There is a critical need for sensors that provide real-time data regarding material evolution under highly accelerated irradiation. These sensors would ideally have a small profile and the ability to withstand irradiation at extremely high dose rates and temperatures for extended periods of time. A capacitance-based sensor is currently under development at the University of Tennessee to provide a direct measurement of in-pile dimensional change during irradiation. Sensor response was simulated using AutoCAD Electromagnetic field simulator (EMS) for a variety of sensor materials and configurations and fuel pin swelling conditions. Initial results of these simulations are summarized and areas of ongoing research and development are discussed.

Naughton, Tyler↗

DOE OSTI2019

High-Frequency Acoustic Mode Identification of Unstable Combustors

Abstract High frequency thermoacoustic instabilities are becoming increasingly problematic in modern combustion systems. Understanding which acoustic mode is being excited is important for understanding potential mechanisms and control approaches — for example, influence of a helical shear layer mode on the flame has profoundly different effects on the first tangential acoustic mode, than a higher order axial mode of similar frequency. Nonetheless, the modal density increases with frequency and it becomes increasingly difficult to determine which acoustic mode is self-excited, based upon frequency calculations alone. Moreover, access issues and cost usually limit the number of pressure probes that can be distributed axially and azimuthally in the combustor. This paper presents a methodology for identifying the acoustic mode by using high temperature pressure transducers flush mounted in a combustion chamber. Modal identification is demonstrated with a siren under non-reacting conditions. The siren is mounted on the chamber to excite longitudinal and azimuthal waves. Five acoustic sensors at different axial and azimuthal locations measure the pressure fluctuations simultaneously. Given the forcing frequency and the speed of sound, the pressure distribution in the combustor is reconstructed in the time domain from the measured data by using a least squares method to determine its mode shapes. In addition, the finite element method (FEM) solver is used to provide the eigenfrequencies and corresponding mode shapes. The test results demonstrate that the mode shapes from the reconstructed data and corresponding frequencies are consistent with those predicted from the FEM, which validates the methodology in this study. In addition, the methodology is extended to practical reacting cases without the siren to determine the acoustic mode shapes of naturally occurring instabilities. In these cases, the modal features have strong stochastic features, such as what appear to be stochastic variations in overall amplitude and relative amplitudes of clockwise and counterclockwise waves.

Kim, J.↗

DOE OSTI2015

Calculation of neutron and gamma fluxes in support to the interpretation of measuring devices irradiated in the core periphery of the OSIRIS Material Testing Reactor

Technological irradiations carried out in material testing reactors (MTRs) are used to study the behavior of materials under irradiation conditions required by different types of nuclear power plants (NPPs). For MTRs, specific instrumentation is required for the experiment monitoring and for the characterization of irradiation conditions, in particular the flux of neutrons and photons. To measure neutron and photon flux in experimental locations, different sensors can be used, such as SPNDs (self-powered neutron detectors), SPGDs (self-powered gamma detectors) and ionization chambers. These sensors involve interactions producing ultimately a measurable electric current. Various sensors have been recently tested in the core periphery of the OSIRIS reactor (located at the CEA-Saclay center) in order to qualify their responses to the neutron and the photon flux. One of the key input data for this qualification is to have a relevant evaluation of neutron and gamma fluxes at the irradiation location. The objective of this work is to evaluate the neutron and the gamma flux in the core periphery of the OSIRIS reactor. With this intention, specific neutron-photonic three-dimensional calculations have been performed and are mainly based on the TRIPOLI-4{sup R} three-dimensional continuous-energy Monte Carlo code, developed by CEA (Saclay Center) and extensively validated against reactor dosimetry benchmarks. In the case of the OSIRIS reactor, TRIPOLI-4{sup R} code has been validated against experimental results based on neutron flux and nuclear heating measurements performed in ex-core and in-core experiments. In this work, simultaneous contribution of neutrons and gamma photons in the core periphery is considered using neutron-photon coupled transport calculations. Contributions of prompt and decay photons have been taken into account for the gamma flux calculation. Specific depletion codes are used upstream to provide the decay-gamma sources required by TRIPOLI-4 calculations. In this paper, the neutron-photonic calculation scheme used for the neutron-gamma flux evaluation in the OSIRIS core periphery will be first detailed. Then, a comparison of calculated and measured values of neutron and gamma fluxes will be introduced. (author)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

DOE OSTI2021

Field Validation of a Smart Energy Recovery Ventilation System Using Low-Cost Indoor Air Quality Sensors

This project is a field validation, using low-cost indoor air quality (IAQ) sensors, of a smart ventilation system that can help low-load homes in humid environments maintain acceptable indoor humidity conditions while providing adequate ventilation according to ASHRAE 62.2. The objectives of this research were to (1) address builders’ concerns with mechanical ventilation in humid environments and (2) answer the question of whether smart control logic helps with occupant comfort and the creation of a more acceptable indoor environment. To address the objectives of the study, the Southface team collected field data for one year in four Charleston, South Carolina, new construction homes in order to determine the differences in occupant comfort; comfort metrics; IAQ; and heating, ventilating, and air-conditioning (HVAC) energy consumption when toggling biweekly between an energy recovery ventilator (ERV) operating continuously and an ERV operating with smart, time-varying humidity control logic. The smart ventilation algorithm under consideration in this field test did create a less humid indoor environment on an annual basis as quantitatively measured through temperature and relative humidity (T/RH) readings, expressed most discernably as “percentage of time above 60% RH” and “percentage of time above 55°F dewpoint.” However, the difference it made was inconsistent during the spring, summer, and fall months, and it was only directionally consistent during the winter months. We suspect that this is primarily due to the long runtimes and concomitant dehumidification activity of the air-conditioning (A/C) units in response to the high sensible loads in Charleston. The effect of the smart ventilation algorithm was not discernable to the occupants in this study, as recorded through seasonal surveys.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2025

Flood Susceptibility Mapping Using Machine Learning and Geospatial-Sentinel-1 SAR Integration for Enhanced Early Warning Systems

This study presents a comprehensive framework for flood susceptibility mapping by integrating geospatial factors with both statistical and machine learning models. Thirteen Flood-related factors, including DEM, slope, TWI, NDVI, etc., are extracted as features of models, and historical flood data derived from Sentinel-1 SAR from 2018 to 2023 are used as the target variables of the models. These datasets are analyzed using a frequency-based statistical model and three machine learning models, including Random Forest, XGBoost, and CNN, to generate flood susceptibility maps. The performance of each model is evaluated through AUC; and SHAP scores are separately generated for Machine learning (ML) models to explain each feature contribution in the ML model. The generated susceptibility maps are validated by high-flood-risk locations monitored by flood sensors, BLE inundation models, and flood-prone areas suggested by the Local Community Task Force. The results indicate that the XGBoost model outperforms all other models, with an AUC of 0.92 and demonstrates the highest alignment with recommended high-flood-risk locations, while the frequency-based statistical model showed the weakest performance with an AUC of 0.65. SHAP value graphs highlight the elevation, slope, and TWI as the most influential features across all models. The susceptibility maps generated by the machine learning model show strong agreement with the BLE map and high-flood-risk areas identified by the local Community Task Force.

Google Engine↗

DOE OSTI2021

Identification of High-Frequency Transverse Acoustic Modes in Multi-Nozzle Can Combustors

Abstract High frequency thermoacoustic instabilities are problematic for lean-premixed gas turbines. Identifying which acoustic mode is being excited is important, in that it provides insight into potential mitigation measures and mechanical stress/life. However, the frequency spacing between modes becomes quite close for high frequency instabilities in a can combustor. This makes it difficult to distinguish between the modes (e.g., the first transverse mode vs. a higher order axial/mixed mode) based upon frequency calculations alone, which inevitably have uncertainties in boundary conditions, temperature profiles, and combustion response. This paper presents a methodology to simultaneously identify the acoustic mode shapes in the axial and azimuthal directions from acoustic pressure measurements. Multiple high temperature pressure transducers, located at distinct axial and azimuthal positions, are flush mounted in the combustor wall. The measured pressure oscillations from each sensor are then used to reconstruct the pressure distributions by using a least squares method in conjunction with a solution of a three dimensional wave equation. In order to validate the methodology, finite element method (FEM) calculations with estimated post-flame temperature is used to provide the candidate frequencies and corresponding mode shapes. The results demonstrate the reconstructed mode shapes and standing/spinning character of transverse waves, as well as the associated frequencies, both of which are consistent with the FEM predictions. Nodal line location was also extracted from the experimental data during the instabilities in the pressure data.

Kim, J.↗

DOE OSTI2022

Strain gauge for testing microreactor hexagonal core blocks in the Single Primary Heat Extraction and Removal Emulator

To support the development and deployment of microreactor technologies, experiments that help verify and validate reactor systems and components are performed at non-nuclear test facilities. The Single Primary Heat Extraction Removal Emulator at Idaho National Laboratory is one of these test facilities and was used to monitor a test article that has a prototypic geometry of a heat-pipe cooled microreactor core block. To collect crucial temperature and strain data during testing, temperature and strain sensing fiber optics were embedded to the surface of the test article using an ultrasonic additive manufacturing technique. To support and provide benchmark strain data for the embedded sensors, a commercial resistive strain gauge was attached. This report will discuss the results from the deployment of the commercial strain gauge which includes the setup/attachment strategies, data acquisition, and analysis of the strain data.

36 MATERIALS SCIENCE↗

DOE OSTI2022

Dry in-pile fracture test (DRIFT) for separate-effects validation of ceramic fuel fracture models

Fracture is an important component of nuclear fuel behavior, and significant efforts have been invested into developing fuel performance models that are capable of accurately representing fracture. Usable data on the process of fracture propagation in nuclear fuel under realistic service conditions are very limited. To address this need, a series of separate-effects experiments were developed and performed at Idaho National Laboratory's Transient Reactor Test (TREAT) facility. These experiments employ a heat sink to radially remove heat from the fuel in a manner that approximates the effect of coolant in an operating light-water reactor (LWR). The test holder for these experiments is known as the Dry In-pile Fracture Test (DRIFT). A series of experiments employing DRIFT and TREAT were performed to provide data on the extent and nature of fracture in fresh fuel at various points during a ramp to full power. Novel aspects of these experiments include the way they employ a heat sink to replicate steady-state LWR conditions, as well as the use of fiber optic sensors for in-reactor thermal instrumentation. Details on the development of this experiment, experimental conditions, and resulting data (including in situ thermal measurements and post-irradiation imaging of fracture) are provided in this work. LWR-equivalent powers ranging from 10 to 25 kW/m were tested using this apparatus. Cracking was visible at all power levels, with increasing cracking extent as the power level increased, although there was little difference in the cracking between the two highest-power tests, which had LWR-equivalent powers of 20 and 25 kW/m.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

DOE OSTI2013

Experimental validation of a high voltage pulse measurement method.

This report describes X-cut lithium niobate s (LiNbO3) utilization for voltage sensing by monitoring the acoustic wave propagation changes through LiNbO3 resulting from applied voltage. Direct current (DC), alternating current (AC) and pulsed voltage signals were applied to the crystal. Voltage induced shift in acoustic wave propagation time scaled quadratically for DC and AC voltages and linearly for pulsed voltages. The measured values ranged from 10 - 273 ps and 189 ps 2 ns for DC and non-DC voltages, respectively. Data suggests LiNbO3 has a frequency sensitive response to voltage. If voltage source error is eliminated through physical modeling from the uncertainty budget, the sensor s U95 estimated combined uncertainty could decrease to ~0.025% for DC, AC, and pulsed voltage measurements.

Cular, Stefan↗

DOE OSTI2024

Automated Assembly Progress Monitoring in Modular Construction Factories Using Computer Vision-Based Instance Segmentation

Modular construction has recently gained interest as a transformative construction method. In this method, a large portion of the construction is performed inside factories, where processes are fast-paced and interdependent; therefore, any deviation from the schedule can delay the production. Such deviations are frequent in modular factories due to the labor-intensive nature of the tasks. This propagation of delays can be mitigated by continuously monitoring each process; however, current manual monitoring methods are laborious, and recently proposed contact sensor-based methods are intrusive to the work. In addition, recent computer vision-based monitoring methods inside factories are limited to detection algorithms that fail to provide the pixel-level accuracy required for assembly progress monitoring in highly occluded factory scenes, and they require a large number of manual annotations. Therefore, this paper proposes a method to monitor the installation of subassemblies in modular construction factories using mask R-CNN instance segmentation and improves the data efficiency of the model using a copy-paste augmentation method. This method was validated on the CCTV videos captured from a modular construction factory in the US, resulting in a 9% mAP improvement in segmentation.

computer vision↗

DOE OSTI2013

Uniform and Half‐Space Models for Magnetic Dipole Sources in a Conducting Medium: I. Theory and Results

Underground nuclear events result in electromagnetic signals near the event location. Of particular interest, because its signal can be potentially enhanced by the use of large cavities to decouple the seismic signal, is the generation of an EMP signal within the nuclear detonation (NUDET) cavity, from the formation of a bubble in the local magnetic field. This mechanism is equivalent to a buried time varying magnetic dipole. In order to model such signals Los Alamos National Laboratory is developing a detailed model for the propagation of RF signals from buried sources to nearby sensors, with a focus on sources that are magnetic dipole in character. In order to validate this code, models for simple geometries involving magnetic dipole sources have been implemented. These models represent the fields from vertical and horizontal dipoles in cylindrical coordinates, either in a uniform linear isotropic electrical medium, or in a space consisting of two half spaces interfacing at the plane z=0, with the upper half space having the electrical properties of free space and the lower half space having arbitrary linear conductivity and electrical permittivity. While primarily intended for the validation of the more detailed model, the shorter turnaround time of these models facilitates the study of the general properties of geomagnetic bubble signals. This report summarizes the theory and example results for those models. The use of the model for model validation is illustrated with the near and surface fields for dipoles with Gaussian time dependence. Also examined are the surface fields expected for the geomagnetic bubble predicted by the numerical model of Kovalenko et al for a 1 kt nuclear detonation. The model of Kovalenko et al was also rescaled to allow comparison with Zablocki’s electric field observations for Bilby, Event III, and Event V. While the rescaled Kovalenko et al model had approximately the right time scale for the duration of the main peak, and a slower fall than rise also in qualitative agreement with the observations, the predicted signals had very different polarizations, were too low by more than an order of magnitude, and had less temporal structure. A simple comparison with the vertical B field data of Sweeney, where the detailed event geometries and yields are classified, but the sensor range and bounds on the yields are given, suggest that the scaled Kovalenko et al model gives too small a peak field compared to observations for that data set as well, except perhaps for the Borate event. A companion report will give a detailed discussion of the implementation and code.

58 GEOSCIENCES↗

DOE OSTI2023

PRIMED for the Future: Purposing Raw Intake for Machine Learning-Enabled Detection (Final Report)

The COVID-19 pandemic demonstrated how a novel, elusive, and diffuse biological threat can engender uncertainty and misinformation, and it underscored the need for flexible analytical modalities agnostic to the identity of biological material. Yet even before the pandemic, recognition of the limitations of the current, list-based approach, which focuses on known pathogens and biotoxins, and of the importance of agent-agnostic biodetection was growing within the biosecurity community. In a 2018 report on “Biodefense in the Age of Synthetic Biology,” for example, the National Academy of Sciences stated that “an overreliance on the Select Agent List is a systematic weakness affecting many aspects of the United States’ current biodefense mitigation capability”. More recently, a group of biodefense researchers proposed the identification and adoption of “bioagent-agnostic signatures (BASs)” as a way of detecting and characterizing not only existing agents but also novel ones, an approach they believe will “enable a more flexible and resilient biodefense posture”. Indeed, the future of biodetection requires us to begin developing novel analytics that can identify anomalies and/or characteristics that indicate a potential threat, whether known or unknown, without looking for a specific signature that has been identified previously. To assess potential threats more rapidly, it is critical to develop agnostic artificial intelligence (AI)/machine learning (ML) systems that can be employed for real-time assessment of the nature and source of a perturbation. Such systems should be multi scale and multi-dimensional, integrating sensor data from a range of biological, chemical, and physical application spaces. Emerging deep learning (DL) models demonstrate exceptional promise for identification of discriminatory features within multi-dimensional datasets. DL models have the capacity to recognize and encode highly complex patterns in a wide range of input data modalities, including images, text, and biological/chemical/physical spectra. As such, they can execute a wide range of assessments and determinations that have traditionally required a human operator. The promise of advances in DL is apparent in the realm of human health and medicine. DL models have been validated for evaluating a variety of clinical threats to human health in a range of contexts, including infection and cancer, and they demonstrated improved performance in predicting stroke relative to human neurologists in some categories of data. Continuously evolving advances in AI/ML are expected to support more efficient evaluation of raw sequence, spectroscopy, and spectrometry data. For instance, recent advances and deployment of large language models (LLM) such as Generative Pre training Transformer (GPT) and Bidirectional Encoder Representations from Transformers (BERT) have already motivated application of these models for biological function prediction. As frameworks such as LLMs become larger and more complex in their representations, their capacity to serve as pre-trained models that can be fine-tuned for biological/biodetection purposes will similarly be amplified. While existing and emerging AI/ML have found broad applicability and use cases in the clinical sciences, development for environmental evaluation and biodetection has been limited. Functionalizing such capabilities for this purpose requires an understanding of the existing technical landscape and how the respective tools and algorithms are currently being employed. This landscape awareness then allows an assessment of the current practical capabilities of existing models and the anticipated requirements and development efforts that will be needed to adapt available algorithms for biodetection applications relevant to DHS. Leveraging expertise in biodetection, ML, and operational biodetection, the effort described in this report is comprised of a systematic landscape assessment (Subtask 2.1), comparative evaluation (Subtask 2.2), and formulation of a value proposition (Subtask 2.3) for the prospect of ML-enabled, agnostic biodetection from raw, or minimally-processed, datasets.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2023

Development of Optical Fiber-Based Sensors for Nuclear Microreactor Structural Health Monitoring

This report provides an experimental assessment of two different optical fiber–based acoustic sensors that are being investigated for application in nuclear microreactors to enhance structural health monitoring capabilities. Optical fibers are resilient in high-temperature and high-radiation environments, have a small sensor footprint, are immune to electromagnetic interference, and are capable of spatially distributed sensing. The two sensors investigated here are (1) Fabry–Pérot Cavities (FPCs) between two copper-coated fibers, embedded in nickel capillary tubes and (2) type-I fiber Bragg grating (FBG) arrays contained within metal capillary tubes. These sensors can be interrogated using low-coherence interferometry or swept wavelength interferometry, respectively, to measure the resonant frequencies of the components or systems to which these sensors are bonded. The FPC developed herein has been subjected to temperatures up to nearly 800°C while tack-welded to a tubular test specimen. Even at the highest temperatures, the measured resonant frequencies compared well with those obtained using an accelerometer that was bonded to an unheated portion of the specimen. The FBG array was tested in multiple bonding configurations to a tubular test specimen, all at room temperature, with the understanding that high temperature (i.e., type II) FBGs could be used to obtain similar data at high temperatures; the FBG array data were validated with noncontact laser vibrometry, which is being used at Los Alamos National Laboratory to relate acoustic signatures to component stresses and/or structural defects. The goal of this work is to identify the most promising techniques that are also compatible with operation in a microreactor environment.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

DOE OSTI2025

New frontiers in wind-wildlife monitoring systems

Effective minimization of negative effects of wind energy on wildlife is an iterative process whereby direct observations of wildlife effects inform and validate mitigation strategies. Yet, the full implementation of this adaptive management has been hindered by a lack of appropriate data. The accurate, high-resolution data required exceeds the capacity of most current monitoring approaches (human observers or monitoring technologies applied in isolation). Current applications of monitoring technologies struggle to harness their full potential by failing to capitalize on opportunities for integration with additional technologies and/or by having limited temporal and spatial resolution. At the emergence of this new frontier of wildlife monitoring, we review the elements of a robust wind-wildlife monitoring system and highlight sensor fusion principles that facilitate effective implementation and integration of multiple monitoring technologies. We also illustrate how sensor fusion solutions can generate high resolution data on collision and displacement effects on terrestrial wildlife across complex spatial and temporal scales.

17 WIND ENERGY↗

DOE OSTI2024

Neuro-Spark: A Submicrosecond Spiking Neural Networks Architecture for In-Sensor Filtering

Neuro-Spark, which is a new neuromorphic architecture with a field-programmable gate array (FPGA) implementation for ultrafast spiking neural network (SNN) inference at the edge, facilitates smart-pixel in-sensor filtering for high-energy physics experiments at the Large Hadron Collider (LHC). Utilizing the evolutionary optimization for neuromorphic systems (EONS) training method, we generate compact SNN models with 91% signal efficiency, akin to convolutional neural networks but with half the parameters. However, deploying near the detector poses a challenge because the SNN must handle a sustained input data rate exceeding 1013 GB/s. To overcome this, we propose a novel hardware architecture that uses high-level synthesis to construct a tuned architecture for the EONS-trained SNN. In addition to the analysis and validation with an AMD Xilinx Artix-A7 FPGA, our solution consumes only ç24% of FPGA LUT and flipflops. We also introduce an innovative quantization method that reduces FPGA resource utilization by ç15% without compromising accuracy. Our FPGA implementation achieves computing latency of ç10 ns for smart-pixel application inference on an edge FPGA.

Miniskar, Narasinga Rao↗

DOE OSTI2016

Authenticated sensor interface device

A system and method for the secure storage and transmission of data is provided. A data aggregate device can be configured to receive secure data from a data source, such as a sensor, and encrypt the secure data using a suitable encryption technique, such as a shared private key technique, a public key encryption technique, a Diffie-Hellman key exchange technique, or other suitable encryption technique. The encrypted secure data can be provided from the data aggregate device to different remote devices over a plurality of segregated or isolated data paths. Each of the isolated data paths can include an optoisolator that is configured to provide one-way transmission of the encrypted secure data from the data aggregate device over the isolated data path. External data can be received through a secure data filter which, by validating the external data, allows for key exchange and other various adjustments from an external source.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2018

Authenticated sensor interface device

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗