SEARCH · Engineering Papers

Results for “anomalous events”

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 73 records · Page 4

All sources NASA NTRS DOE OSTI

DOE OSTI2022

Online-compatible unsupervised nonresonant anomaly detection

There is a growing need for anomaly detection methods that can broaden the search for new particles in a model-agnostic manner. Most proposals for new methods focus exclusively on signal sensitivity. However, it is not enough to select anomalous events—there must also be a strategy to provide context to the selected events. In this work, we propose the first complete strategy for unsupervised detection of nonresonant anomalies that includes both signal sensitivity and a data-driven method for background estimation. Our technique is built out of two simultaneously trained autoencoders that are forced to be decorrelated from each other. This method can be deployed off-line for nonresonant anomaly detection and is also the first complete on-line-compatible anomaly detection strategy. We show that our method achieves excellent performance on a variety of signals prepared for the ADC2021 data challenge.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2025

Anomaly Detection in Electronic Health Records Across Hospital Networks: Integrating Machine Learning With Graph Algorithms

In a large hospital system, a network of hospitals relies on electronic health records (EHRs) to make informed decisions regarding their patients in various clinical domains. Consequently, the dependability of the health information technology (HIT) systems responsible for collecting EHR data is of utmost importance for patient safety. Recently, novel methods and tools aimed at identifying anomalies in EHR data to bolster the reliability of HIT systems have been introduced. However, these existing methods and tools primarily concentrate on individual hospitals, which limits our understanding of system-wide anomalous events and their potential impact on patient safety across multiple hospitals. In this article, we introduce a new approach to detecting anomalies in EHR data within a network of hospitals. This is achieved by combining advanced machine learning techniques with graph algorithms to create a tool capable of swiftly identifying and responding to deviations. Our proposed approach employs a combination of five machine learning models, harnessing the unique strengths of each model to provide a more robust detection system. The detected anomalies are then represented as graphs, allowing us to recognize patterns across the hospital network. This aids in identifying anomalies that span multiple medical facilities, potentially indicating broader system-level risks. Extensive real-world testing of our approach demonstrated its ability to offer actionable insights compared to existing methods. Additionally, its scalable design ensures seamless integration into existing HIT infrastructures.

Niu, Haoran [Oak Ridge National Laboratory (ORNL),↗

DOE OSTI2022

Network Anomaly Detection Using Federated Learning

The internet is turning out to be an integral part of every-one's lives as more and more devices are being connected to serve societal needs. Our work is motivated by two ma-jor observations. Firstly, one drawback of connecting to the network is the threat of network attacks that can compromise users' private information, leading to data loss and adversely affecting productivity. There are several traditional security mechanisms to defend against these attacks, such as firewalls, virtual private networks (VPNs), demilitarized zones (DMZs), and vulnerability scanners. One way to prevent these attacks is early detection and prevention. However, these kinds of architecture do not scale very well because of their centralized nature. Secondly, we observe from heuristics and data set distributions that the majority of the requests made to a server are innocuous. Therefore, almost all server request data sets are highly imbalanced, weighted highly towards the harmless requests.

Marfo, William↗

DOE OSTI2021

Intelliquench: An Adaptive Machine Learning System for Detection of Superconducting Magnet Quenches

In superconducting magnets, the irreversible transition of a portion of the conductor to resistive state is called a “quench.” Having large stored energy, magnets can be damaged by quenches due to localized heating, high voltage, or large force transients. Unfortunately, current quench protection systems can only detect a quench after it happens, and mitigating risks in Low Temperature Superconducting (LTS) accelerator magnets often requires fast response (down to ms). Additionally, protection of High Temperature Superconducting (HTS) magnets is still suffering from prohibitively slow quench detection. In this study, we lay the groundwork for a quench prediction system using an auto-encoder fully-connected deep neural network. After dynamically trained with data features extracted from acoustic sensors around the magnet, the system detects anomalous events seconds before the quench in most of our data. While the exact nature of the events is under investigation, we show that the system can “forecast” a quench before it happens under magnet training conditions through a randomized experiment. This opens up the way of integrated data processing, potentially leading to faster and better diagnostics and detection of magnet quenches

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

DOE OSTI2021

Machine Learning Guided Operational Intelligence from Synchrophasors (Final Report)

Schweitzer Engineering Laboratories (SEL) and Oregon State University (OSU) received over 27 terabytes of electrical power system phasor measurement unit (PMU) data for the Eastern, Western, and ERCOT interconnections. The dataset includes measurements spread across 446 PMUs from early 2016 to mid 2018 depending on the interconnect. The full dataset was split into a training and test (holdout) dataset by PNNL. All data was received in the Apache Parquet format. The overarching goal of this project is to develop and execute a strategy to mitigate data anomalies, perform analysis on the dataset, and detect anomalous events in the data.

24 POWER TRANSMISSION AND DISTRIBUTION↗

DOE OSTI2021

Department of Energy Water Power Technologies Office Cyber Response & Recovery Flipbook [Slides]

Protecting hydroelectric plants from incidents that adversely impact their cyber-physical systems presents unique challenges due to the plants’ widely dispersed geographic locations and varied configurations as well as the relative nascent nature of the cyberattacks targeting these facilities. To help hydroelectric plants better respond to and mitigate cybersecurity incidents, this Department of Energy Water Power Technologies Office Cyber Response & Recovery Flipbook is to be used at a hydroelectric plant to quickly respond to an anomalous event. In addition to this product, there are three other products meant to be distributed to a hydroelectric plant to assist in their cyber incident response and recovery. The first, a report on the processes of building this flip book based on a large set of existing guidance. The second, a handy guide of hydroelectric and cyber guidance in responding to the cyber and physical systems within a hydroelectric plant. And the third is a correlated alignment of the steps an hydroelectric plant operator would take for both a cyber incident as well as an emergency response process if the event rises to a cyber incident affecting the safe and reliable operations of a hydroelectric plant.

13 HYDRO ENERGY↗

DOE OSTI2020

Hydroelectric Cybersecurity Response and Recovery Overview

Protecting hydroelectric plants from incidents that adversely impact their cyber-physical systems presents unique challenges due to the plants’ widely dispersed geographic locations and varied configurations as well as the relative nascent nature of the cyberattacks targeting these facilities. To help hydroelectric plants better respond to and mitigate cybersecurity incidents, this Department of Energy Water Power Technologies Office Cyber Response & Recovery Overview document discusses the process of defining how a hydroelectric plant might respond to and recover from an anomalous event. In addition to this product, there are three other products meant to be distributed to a hydroelectric plant to assist in their cyber incident response and recovery. The first, a handy flip book that guides an operator in the midst of a cyber event through the R&R process of the incident and if the event warrants, through an emergency action plan to recover the plant itself. The second, a handy guide of hydroelectric and cyber guidance in responding to the cyber and physical systems within a hydroelectric plant. And the third is a correlated alignment of the steps an hydroelectric plant operator would take for both a cyber incident as well as an emergency response process if the event rises to a cyber incident affecting the safe and reliable operations of a hydroelectric plant.

13 HYDRO ENERGY↗

DOE OSTI2023

Investigating Anomalies in Compute Clusters: An Unsupervised Learning Approach

As compute clusters continue to grow in scale and complexity, the frequency of detected anomalies in their operation significantly increases. Timely detection of anomalous events is vital to maintain system efficiency and availability. This study presents an attentionbased graph neural network (GNN) for detecting anomalies in clusters at the compute node level and for providing detailed root cause analysis. We show the effectiveness of attention-based GNNs to accurately detect and localize anomalies on real-world datasets.

Lu, Yiyang↗

NASA NTRS1976

The Skylab barium plasma injection experiments. II - Evidence for a double layer

Television observations of a barium-plasma flux tube extending from near 4500 km to near 10,000 km during a magnetic substorm and dawn-sector auroral display indicated several interesting anomalous events. Beyond 5500 km, there was a rapid increase in brightness accompanied by flux-tube splitting and diffusion, leaving behind a truncated single flux tube. From the orientation of the flux tube compared with theoretical field models, the presence of a substantial field-aligned current sheet is deduced. A suggested explanation of these phenomena is given in terms of a plasma potential double layer.

Wescott, E. M.↗

NASA NTRS1977

Spacecraft charging anomalies on the DSCS II, launch 2 satellites

The six different types of anomalous events occurred on two DSCS II satellites. The total number of events, over 100, and the long operational period, nearly 3 years, permits some statistical analyses to be performed. Correlation of occurrences of particular types of anomalies with equinoxes and seasons of the year are consistent with a spacecraft charging model. On the other hand, an interesting correlation of occurrences with days of the week has been found. Finally, a long term diminution in the frequency of occurrence of events has been observed, and is discussed in terms of environmental activity, material degradations and the need for more data.

Inouye, G. T.↗

NASA NTRS1979

On the approach to forecasting polar ionospheric conditions

The major properties of polar ionospheric main anomalous events are summarized. The monitoring of large scale features of the ionization distribution that are the projections of large scale structural characteristics of magnetospheric plasma on the upper ionosphere is suggested as a basic principle of polar ionospheric condition forecasting. It is concluded that the processes of the magnetosphere/ionosphere interaction appear to play a predominant role in the creation of the polar ionosphere.

Besprozvannaya, A. S.↗

NASA NTRS1982

Anomalous behavior of the Pioneer Venus entry probes during lower descent

Four Pioneer Venus Probe spacecraft entered the Venusian atmosphere on December 9 1978, and descended to the surface while making in situ measurements of chemical composition, structure, and radiative balance. In the lower Venusian atmosphere, a number of anomalous events were noted in some of the scientific instruments and Probe engineering data. Many of these anomalies occurred on each of the Probes at approximately the same altitude. The anomalies are thought to be the result of some unexpected electrical interaction between the Probe and the Venusian atmosphere. This conclusion is based on analysis and testing of similar Probe hardware on the ground. However, the possibility that some anomalies were a result of latent design of manufacturing flaws cannot be ruled out.

Polaski, L. J.↗

NASA NTRS1988

Rocket temperature soundings

In situ rocket-borne measurements of temperature and wind contribute to a better determination and understanding of stratospheric behavior and, hence, to a better understanding of processes that control the dynamical and chemical behavior of this region. Concern over ozone depletion and the difficulty generally involved in determining actual ozone trends has generated significant interest in temperature behavior, especially trends. Recent analysis of rocketsonde acquired temperature data between 1969 and 1986 contains evidence that the stratosphere may indeed be cooling. It is the intention of the RTOP to continue to provide rocketsonde measurements to: maintain the long term data stratospheric-mesospheric data base already established for Wallops; provide ground truth for remote measurements; and continue studies of atmospheric structure and morphology of disturbances and anomalous events as resources permit.

Schmidlin, F. J.↗

NASA NTRS1988

Plans for the extreme ultraviolet explorer data base

The paper presents an approach for storage and fast access to data that will be obtained by the Extreme Ultraviolet Explorer (EUVE), a satellite payload scheduled for launch in 1991. The EUVE telescopes will be operated remotely from the EUVE Science Operation Center (SOC) located at the University of California, Berkeley. The EUVE science payload consists of three scanning telescope carrying out an all-sky survey in the 80-800 A spectral region and a Deep Survey/Spectrometer telescope performing a deep survey in the 80-250 A spectral region. Guest Observers will remotely access the EUVE spectrometer database at the SOC. The EUVE database will consist of about 2 X 10 to the 10th bytes of information in a very compact form, very similar to the raw telemetry data. A history file will be built concurrently giving telescope parameters, command history, attitude summaries, engineering summaries, anomalous events, and ephemeris summaries.

Marshall, Herman L.↗

NASA NTRS1990

A new model for the calculation and prediction of solar proton fluences

A new predictive engineering model for the energy greater than 10 MeV and greater than 30 MeV solar proton environment at earth is reviewed. The data used are from observations made from 1956 through 1985. In this data set, the distinction between 'ordinary events' and 'anomalously large events' that was required in earlier models disappeared. This permitted the use of statistical analysis methods developed for ordinary events on the entire data set. The greater than 10-MeV fluences with the new model are about twice those expected on the basis of earlier models. At energies greater than 30 MeV, the old and new models agree.

Feynman, Joan↗

NASA NTRS1990

New interplanetary proton fluence model

A new predictive engineering model for the interplanetary fluence of protons with above 10 MeV and above 30 MeV is described. The data set used is a combination of observations made from the earth's surface and from above the atmosphere between 1956 and 1963 and observations made from spacecraft in the vicinity of earth between 1963 and 1985. The data cover a time period three times as long as the period used in earlier models. With the use of this data set the distinction between 'ordinary proton events' and 'anomalously large events' made in earlier work disappears. This permitted the use of statistical analysis methods developed for 'ordinary events' on the entire data set. The greater than 10 MeV fluences at 1 AU calculated with the new model are about twice those expected on the basis of models now in use. At energies above 30 MeV, the old and new models agree. In contrast to earlier models, the results do not depend critically on the fluence from any one event and are independent of sunspot number. Mission probability curves derived from the fluence distribution are presented.

Feynman, Joan↗

NASA NTRS1993

The physical mechanism of comet outbursts: An experiment

During a series of impact experiments into regolith-like powders at the NASA Ames Research Center Vertical Gun Facility in 1976, I observed and filmed a unique anomalous event that may illuminate outburst mechanisms in comets. During one test, a new batch of basalt powder (half the mass in particles less than 800 microns in diameter) retained some air pressure while the vacuum chamber was being evacuated. As a result, the projectile impacted into gas-charged regolith. Instead of ejecting the normal, relatively negligible amount of debris, the disturbance triggered a major eruption that lasted at least 18 seconds. The experimental results have been recently re-analyzed with reference to cometary phenomena. A series of frames from this eruption experiment are shown. The ejecta velocities of 150 to 300 cm/s would have been sufficient to drive debris into the coma of a comet nucleus smaller than a few kilometers diameter. The event suggests a mechanism for comet outbursts, discussed briefly by Hartmann et al.: the pore space in a layer of regolith, possibly with weak effective tensile strength, becomes gas charged as ice slowly sublimates. Once the effective tensile strength is exceeded by the gas pressure, the surface fails locally, triggering an eruption such as photographed here. This model is consistent with the emerging view of regolith materials on comets and is closest to the recent model of Rickman et al. The earlier models generally picture a more uniform flow of debris off the comet, not outbursts. Rickman et al. allow gas pressure to build until it matches the overburden pressure, followed by 'instantaneous blow-off'. They note that as soon as the mantle is found to be unstable, we consider it to be instantaneously swept away by the gas pressure. The main new points made here are that the experiment gives a more realistic view of the blow-off process after surface failure occurs, and the present model gives a recharge mechanism that can explain recurrent outbursts on comets such as P/Schwassmann-Wachmann 1 and 2060 Chiron. In fact, the resulting jets resemble distinct jet structures in high-resolution comet comae.

Hartmann, William K.↗

NASA NTRS1995

Pioneer Venus 12.5 km Anomaly Workshop Report, volume 1

A workshop was convened at Ames Research Center on September 28 and 29, 1993, to address the unexplained electrical anomalies experienced in December 1978 by the four Pioneer Venus probes below a Venus altitude of 12.5 km. These anomalies caused the loss of valuable data in the deep atmosphere, and, if their cause were to remain unexplained, could reoccur on future Venus missions. The workshop participants reviewed the evidence and studied all identified mechanisms that could consistently account for all observed anomalies. Both hardware problems and atmospheric interactions were considered. Based on a workshop recommendation, subsequent testing identified the cause as being an insulation failure of the external harness. All anomalous events are now explained.

Seiff, A.↗