Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Anomaly Detection”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

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

At least 541 records · Page 30

Detecting Process Equipment Failures Using Acoustic Data and Machine Learning

Nuclear power plant (NPP) process equipment such as fans, motors, valves, and pumps generate frequent or continuous noise, and deviations from the normal operational sounds made by this equipment can indicate potential issues. These deviations can be identified via automated acoustic anomaly detection, which involves using acoustic sensors (i.e., microphones) alongside detection algorithms to continuously monitor for changes in acoustic signatures. This task is made challenging by the substantial background noise that exists, such as operators opening and closing doors, manipulating valves, and conversing—in addition to typical plant noises. In collaboration with a nuclear power utility partner, this effort assessed the efficacy of acoustic anomaly detection when using a specific acoustic sensor that compresses data into a fixed set of features that are transferable over a standard Internet of Things communication protocol, thereby improving usability but potentially degrading detection performance. Two methods of performing automated acoustic anomaly detection were evaluated: one-class support vector machine (OC-SVM) and isolation forest (iForest). To enable the use of high-quality acoustic data encompassing both normal and anomalous conditions, the study utilized the publicly available Malfunctioning Industrial Machine Investigation and Inspection dataset, which includes real measured acoustic sensor data for a range of equipment types, model numbers, and signal-to-noise ratios (SNRs), along with a benchmark set of detection results. Using this dataset, the methods were tested and then compared against the benchmark results. The results indicated that although the specific acoustic sensor did not enable as rich a feature set extraction, the proposed methods with the limited feature set performed just as well. This provides solid justification for both the methods and the use of the proposed acoustic sensor.

46 - INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AN↗

Improving Variational Autoencoders for New Physics Detection at the LHC With Normalizing Flows

We investigate how to improve new physics detection strategies exploiting variational autoencoders and normalizing flows for anomaly detection at the Large Hadron Collider. As a working example, we consider the DarkMachines challenge dataset. We show how different design choices (e.g., event representations, anomaly score definitions, network architectures) affect the result on specific benchmark new physics models. Once a baseline is established, we discuss how to improve the anomaly detection accuracy by exploiting normalizing flow layers in the latent space of the variational autoencoder.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

ISHM Anomaly Lexicon for Rocket Test

Integrated Systems Health Management (ISHM) is a comprehensive capability. An ISHM system must detect anomalies, identify causes of such anomalies, predict future anomalies, help identify consequences of anomalies for example, suggested mitigation steps. The system should also provide users with appropriate navigation tools to facilitate the flow of information into and out of the ISHM system. Central to the ability of the ISHM to detect anomalies is a clearly defined catalog of anomalies. Further, this lexicon of anomalies must be organized in ways that make it accessible to a suite of tools used to manage the data, information and knowledge (DIaK) associated with a system. In particular, it is critical to ensure that there is optimal mapping between target anomalies and the algorithms associated with their detection. During the early development of our ISHM architecture and approach, it became clear that a lexicon of anomalies would be important to the development of critical anomaly detection algorithms. In our work in the rocket engine test environment at John C. Stennis Space Center, we have access to a repository of discrepancy reports (DRs) that are generated in response to squawks identified during post-test data analysis. The DR is the tool used to document anomalies and the methods used to resolve the issue. These DRs have been generated for many different tests and for all test stands. The result is that they represent a comprehensive summary of the anomalies associated with rocket engine testing. Fig. 1 illustrates some of the data that can be extracted from a DR. Such information includes affected transducer channels, narrative description of the observed anomaly, and the steps used to correct the problem. The primary goal of the anomaly lexicon development efforts we have undertaken is to create a lexicon that could be used in support of an associated health assessment database system (HADS) co-development effort. There are a number of significant byproducts of the anomaly lexicon compilation effort. For example, (1) Allows determination of the frequency distribution of anomalies to help identify those with the potential for high return on investment if included in automated detection as part of an ISHM system, (2) Availability of a regular lexicon could provide the base anomaly name choices to help maintain consistency in the DR collection process, and (3) Although developed for the rocket engine test environment, most of the anomalies are not specific to rocket testing, and thus can be reused in other applications.

Schmalzel, John L.↗

Capturing Complex Multivariate Time Series Interactions to Detect High-Risk Adverse Events During Flight

The reduction of aviation safety metrics below target thresholds continue to drive down the number of aviation fatalities and accidents. To meet future safety demands, sustained efforts by aviation agencies promoting safety assurance processes and systems have prompted ongoing research on identifying and mitigating in-flight risks. With the projected increase in passenger load factor and rollout of more autonomous systems into the national airspace, the need to detect high-risk events in-time or ahead-of-time is becoming increasingly crucial. New anomaly detection and precursor identification algorithms will need to scale to different airframes, levels of autonomy, and system complexity. While the pervasiveness of deep learning has resulted in the development of performant anomaly detection methods, these sophisticated models currently suffer from low end-user interpretability. Building off our previous work on identifying adverse events in multivariate flight data during descent, we propose a data-driven approach for detecting in-flight adverse events caused by the complex interplay of flight variables. Our approach utilizes ordinal patterns of important aircraft stability variables (e.g., airspeed and descent rate) to capture multivariate flight dynamics that can be used to predict the onset of unstable approaches, a high-risk adverse event that can occur during approach. Through the use of ordinal patterns, we aim to create more interpretable detection models of in-flight adverse events that can be translated to future autonomous systems without difficulty. Our analysis shows the presence of distinct ordinal pattern distributions that can be used to predict unstable approaches 1 minute ahead of time with an accuracy of 0.69 and a recall of 0.73 and 30 seconds ahead with an accuracy of 0.70 and a recall of 0.86.

Risk detection↗

Capturing Multivariate Time Series Interactions to Detect High‑Risk Instability During Approach

The reduction of aviation safety metrics below target thresholds continue to drive down the number of aviation fatalities and accidents. To meet future safety demands, sustained efforts by aviation agencies promoting safety assurance processes and systems have prompted ongoing research on identifying and mitigating in-flight risks. With the projected increase in passenger load factor and rollout of more autonomous systems into the national airspace, the need to detect high-risk events in-time or ahead-of-time is becoming increasingly crucial. New anomaly detection and precursor identification algorithms will need to scale to different airframes, levels of autonomy, and system complexity. While the pervasiveness of deep learning has resulted in the development of performant anomaly detection methods, these sophisticated models currently suffer from low end-user interpretability. Building off our previous work on identifying adverse events in multivariate flight data during descent, we propose a data-driven approach for detecting in-flight adverse events caused by the complex interplay of flight variables. Our approach utilizes ordinal patterns of important aircraft stability variables (e.g., airspeed and descent rate) to capture multivariate flight dynamics that can be used to predict the onset of unstable approaches, a high-risk adverse event that can occur during approach. Through the use of ordinal patterns, we aim to create more interpretable detection models of in-flight adverse events that can be translated to future autonomous systems without difficulty. Our analysis shows the presence of distinct ordinal pattern distributions that can be used to predict unstable approaches 1 minute ahead of time with an accuracy of 0.69 and a recall of 0.73 and 30 seconds ahead with an accuracy of 0.70 and a recall of 0.86.

Risk detection↗

Future SDN-HPON Control Plane Architecture and Protocol for On-Demand Terabit End-to-End Extreme-Scale Science Applications

The UC Davis team conducted architecture, algorithm and experimental studies for next generation ultra-high-bandwidth optical networks in support of extreme-scale science applications. In particular, over the course of this 3-year project, the UC Davis team achieved the following main accomplishments: Designed a network architecture for automated multi-domain service provisioning: Designed information exchange schemes between autonomous systems (domains) Designed application programming interfaces to orchestrate resource reservation in data centers, high performance computation facilities, and communication networks belonging to different autonomous systems Experimentally assessed the multi-domain network architecture in a distributed field trial set-up connecting premises in three continents Designed a broker-based orchestrated multi-domain optical networking framework supporting observe-analyze-act cycle based cognitive inter-domain service provisioning: Designed an inter-domain operation mechanism, e.g., topology abstraction, for the broker plane and domain managers to collaboratively accomplish inter-domain service provisioning while respecting the autonomy of each administrative domain Designed an alien wavelength monitoring scheme and an artificial neural network based optical signal-to-noise ratio estimator for quality-of-transmission aware inter-domain service provisioning Developed a deep neural network based multi-domain traffic predictor and a cognitive routing and spectrum assignment algorithm to realize knowledge-based autonomous traffic engineering in multi-domain networks Implemented an optical signal-to-noise ratio estimator and traffic predictor with Google’s Tensor flow and assessed their performance using experimental and synthetic data Demonstrated the benefits of the proposed cognitive schemes with experiments ona two-domain elastic optical network testbed and evaluations on a six-domain simulation framework Built a two-domain software-defined networking control plane testbed using the ONOS platform and demonstrated the workflow of broker and domain manager cooperation, including hierarchical performance monitoring, topology abstraction and cognitive resource allocation. Designed self-learning cognitive service provisioning schemes using advanced machine learning technologies: Designed a self-taught anomaly detection framework for optical networks, including an unsupervised data clustering module for pattern recognition, a supervised data regression and classification module for online anomaly detection, and the system operation principle Experimentally validated the effectiveness of the self-taught anomaly detection framework with various failure/anomaly scenarios. Designed a self-learning autonomic routing, modulation and spectrum assignment (RMSA) provisioning framework for elastic optical networks with deep reinforcement learning. Designed effective modeling and training mechanisms for the self-learning RMSA framework Conducted extensive numerical simulations with various traffic patterns and topologies to verify the advantage of the self-learning RMSA framework.

42 ENGINEERING↗

On the Limits of EM Based Detection of Control Logic Injection Attacks In Noisy Environments

The difficulty in applying traditional security mechanisms in Industrial Control System (ICS) environments makes a large portion of these mission-critical assets vulnerable to cyber attacks. Therefore, there is a dire need for the development of novel security mechanisms specifically designed to protect such critical systems. Recently a lot of attention has been given to mechanisms that exploit the EM emanations of devices for defense purposes. Such practices may lead to the development of robust external and non-intrusive anomaly detection systems. Nevertheless, the majority of current work in the area neglects to consider the implications of real-life environments, particularly environmental noise. In this work, we explore the limits of EM-based anomaly detection towards identifying injection attacks in control logic software in noisy environments. Our study conducted upon both synthetically generated and real signals identified that indeed environmental noise might significantly degrade the accuracy of the anomaly detection process. Experiments done upon synthetic data indicated that assuming that signals are captured with high sampling rates, even minor code injections can be detected with above-90% accuracy in noisy environments where SNR is up to -2dB. This is true even if naive detection methods are considered. Moreover, experiments done using a real-life testbed attest that even single-instruction injections can be detected with near-perfect accuracy in relatively clean environments. Finally, noise-elimination techniques can drastically improve the reliability of the detection mechanism even in noisy environments.

97 MATHEMATICS AND COMPUTING↗

A Visual Analytics Framework for the Detection of Anomalous Call Stack Trees in High Performance Computing Applications

Anomalous runtime behavior detection is one of the most important tasks for performance diagnosis in High Performance Computing (HPC). Most of the existing methods find anomalous executions based on the properties of individual functions, such as execution time. However, it is insufficient to identify abnormal behavior without taking into account the context of the executions, such as the invocations of children functions and the communications with other HPC nodes. We improve upon the existing anomaly detection approaches by utilizing the call stack structures of the executions, which record rich temporal and contextual information. With our call stack tree (CSTree) representation of the executions, we formulate the anomaly detection problem as finding anomalous tree structures in a call stack forest. The CSTrees are converted to vector representations using our proposed stack2vec embedding. Structural and temporal visualizations of CSTrees are provided to support users in the identification and verification of the anomalies during an active anomaly detection process. Furthermore, three case studies of real-world HPC applications demonstrate the capabilities of our approach.

97 MATHEMATICS AND COMPUTING↗

Development of Gamma Background Radiation Digital Twin with Machine Learning Algorithms: Application of Unsupervised Machine Learning to Detection of Anomalies and Nuisances in Gamma Background Radiation Environmental Screening Data

Environmental screening of gamma radiation consists of detecting weak nuisance and anomaly signal in the presence of strong and highly varying background. In a typical scenario, a mobile detector-spectrometer continuously measures gamma radiation spectra in short, e.g., one-second, signal acquisition intervals. The measurement data is a 2D matrix, where one dimension is gamma ray energy, and the other dimension is the number of measurements or total time. In principle, gamma radiation sources can be detected and identified from the measured data by their unique spectral lines. Detecting sources from data measured in a search scenario is difficult due to the highly varying background because of naturally occurring radioactive material (NORM), and low signal-to-noise ratio (S/N) of spectral signal measured during one-second acquisition intervals. The objective of this work is to explore unsupervised machine learning (ML) algorithms for development of a digital twin of gamma radiation background, and for detection and identification of weak nuisances and anomalies events in the presence of highly fluctuating background. In one segment of work, we developed a gamma background estimation model using a Longshort term memory (LSTM) network for one-step CPS time series prediction. The LSTM model was validated with two data sets of measurements from two independent NaI detectors positioned on a mobile platform. The data sets contained background radiation only and no orphan isotope sources. The LSTM model was constructed and tested using data from one of the detectors. Performance of the LSTM model was validate through one-step prediction of CPS time series of another NaI detector without re-training. This approach allows to create a digital twin for nuclear background estimation. Using LSTM, it could be possible to detect a source through subtraction of the estimated counts from the measured background. In another segment of work, we investigated detection of gamma emitting sources in the presence of complex background using unsupervised machine learning. Spectral lines of isotopes are difficult to observe in one-second measurements. Averaging over the entire measurement campaign data set reveals spectral lines of most common background isotopes. Spectral lines of orphan sources, which might appear only in a few measurements during the campaign, will be washed out if averaging is performed over the entire measurement data set. The approach we have explored consists of extracting one-second measurements containing weak spectral features through data clustering. Averaging one-second spectra in a cluster should reveal the presence of anomaly sources. We created two ML models using K-means clustering and Neural Network Self-organizing Map (SOM). Performance of these ML models was benchmarked using search data. One data set contained 137 Cs source, and another dataset contained 131 I source.

54 ENVIRONMENTAL SCIENCES↗

System Engineers and Decisions: It?s All about Knowledge

In order to guarantee that a system meets adequate levels of reliability and availability, system performances are continuously monitored and analyzed thanks to the technological advancements driving the Industry 4.0 revolution. An Industry 4.0 approach is typically based on advanced statistical, big data mining, machine learning, and internet-of-things methods designed to detect anomalies in the behavior of system, detect the most likely failure modes, and provide indications to system engineers on when maintenance activities should be performed before system performance are deemed unacceptable (which can be generated by diagnostic and prognostic methods). However, these analyses, which are designed to automatize and increase the efficacy of the system maintenance program, require large amount of data which can come in various forms: numeric, textual, images, sounds etc. Such data constitutes the historic knowledge benchmark to track system performances and support system engineer decisions. Here we claim that data is not sufficient to support this kind of analyses when applied to systems characterized by complex architectures and behaviors. Robust system engineer decisions require the ability to understand the system operational context that lies behind the observed data elements. In this respect, system models are in fact necessary to “put data in context” and capture relationships between data elements. Industry 4.0 methods require in fact contextual knowledge as a basis upon which hypotheses can be generated and assumptions tested. In our view, for complex systems, model-based system engineering (MBSE) models can afford this contextual knowledge, as they are typically used to describe systems architecture and dynamic behaviors. System knowledge is here intended as the blending of collected data and system architecture which takes the form of a “knowledge graph”. A knowledge graph is a database which consists of a large set of nodes (in our case an entity can be either a data or an MBSE element) which are linked to each other. The types of nodes and links follow a pre-defined topology, sometimes also refers as an ontology, that is designed to fit the actual decisions that needs to be performed. We show here how a knowledge graph can be defined to support system engineer maintenance decisions and how the same graph can be built based on system MBSE models and pre-processed data from numeric (through anomaly detections and diagnostic methods) and textual elements (through technical language processing TLP).

97 - MATHEMATICS AND COMPUTING↗

Robust errant beam prognostics with conditional modeling for particle accelerators

Abstract Particle accelerators are complex and comprise thousands of components, with many pieces of equipment running at their peak power. Consequently, they can fault and abort operations for numerous reasons, lowering efficiency and science output. To avoid these faults, we apply anomaly detection techniques to predict unusual behavior and perform preemptive actions to improve the total availability. Supervised machine learning (ML) techniques such as siamese neural network models can outperform the often-used unsupervised or semi-supervised approaches for anomaly detection by leveraging the label information. One of the challenges specific to anomaly detection for particle accelerators is the data’s variability due to accelerator configuration changes within a production run of several months. ML models fail at providing accurate predictions when data changes due to changes in the configuration. To address this challenge, we include the configuration settings into our models and training to improve the results. Beam configurations are used as a conditional input for the model to learn any cross-correlation between the data from different conditions and retain its performance. We employ conditional siamese neural network (CSNN) models and conditional variational auto encoder (CVAE) models to predict errant beam pulses at the spallation neutron source under different system configurations and compare their performance. We demonstrate that CSNNs outperform CVAEs in our application.

43 PARTICLE ACCELERATORS↗

Autoencoder Based Analysis of RF Parameters in the Fermilab Low Energy Linac

Machine learning (ML) has the potential for significant impact on the modeling, operation, and control of particle accelerators due to its ability to model nonlinear behavior, interpolate on complicated surfaces, and adapt to system changes over time. Anomaly detection in particular has been highlighted as an area where ML can significantly impact the operation of accelerators. These algorithms work by identifying subtle behaviors of key variables prior to negative events. Efforts to apply ML to anomaly detection have largely focused on subsystems such as RF cavities, superconducting magnets, and losses in rings. However, dedicated efforts to understand how to apply ML for anomaly detection in linear accelerators have been limited. In this paper the use of autoencoders is explored to identify anomalous behavior in measured data from the Fermilab low-energy linear accelerator.

Edelen, Jonathan P. (ORCID:0000000215180652)↗

Astronomaly at scale: searching for anomalies amongst 4 million galaxies

ABSTRACT Modern astronomical surveys are producing data sets of unprecedented size and richness, increasing the potential for high-impact scientific discovery. This possibility, coupled with the challenge of exploring a large number of sources, has led to the development of novel machine-learning-based anomaly detection approaches, such as astronomaly. For the first time, we test the scalability of astronomaly by applying it to almost 4 million images of galaxies from the Dark Energy Camera Legacy Survey. We use a trained deep learning algorithm to learn useful representations of the images and pass these to the anomaly detection algorithm isolation forest, coupled with astronomaly’s active learning method, to discover interesting sources. We find that data selection criteria have a significant impact on the trade-off between finding rare sources such as strong lenses and introducing artefacts into the data set. We demonstrate that active learning is required to identify the most interesting sources and reduce artefacts, while anomaly detection methods alone are insufficient. Using astronomaly, we find 1635 anomalies among the top 2000 sources in the data set after applying active learning, including eight strong gravitational lens candidates, 1609 galaxy merger candidates, and 18 previously unidentified sources exhibiting highly unusual morphology. Our results show that by leveraging the human–machine interface, astronomaly is able to rapidly identify sources of scientific interest even in large data sets.

Astronomy & Astrophysics↗

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↗

Comparison of Supervised and Un-Supervised Machine Learning Algorithms for Threat Detection and Scintillator Performance for Radiation Portal Monitoring

Following the events of September 11, 2001, international border crossing have been equipped with radiation portal monitors (RPMs) to identify illicit radioactive material. Polyvinyl toluene (PVT) scintillators are commonly used due to their low cost and reasonable maintainability, however they offer low spectral resolution. Despite the fact that over twenty years has transpired since this event, radioisotopes are still typically identified by hand-crafted classification algorithms, e.g., total counts or energy windowing, and exhibit relatively poor performance in detecting threats at the low false alarm rates required to support the stream of commerce. While some improvement to performance has been realized via the use of supervised machine learning, these classification algorithms typically utilize simulations in lieu of real data due to the sparsity of data for one or more classes. Accordingly, the performance of these algorithms is somewhat less than optimal when examining experiments or simulations with model mismatch. Consequently, in this work, we examine the application of a number of unsupervised machine learning, anomaly detection based algorithms, to circumvent the inverse crime when analyzing spectroscopy data for RPMs. We also compare anomaly detection results with those obtained via the use of supervised classification detection ML algorithms when model mismatch is introduced between the simulated threat items utilized for training/testing. Finally, we compared the performance of the PVT scintillators to those obtained with higher resolution detectors using both anomaly detection and supervised classification algorithms.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗

Time Series Discord Detection in Medical Data using a Parallel Relational Database [PowerPoint]

Recent advances in sensor technology have made continuous real-time health monitoring available in both hospital and non-hospital settings. Since data collected from high frequency medical sensors includes a huge amount of data, storing and processing continuous medical data is an emerging big data area. Especially detecting anomaly in real time is important for patients’ emergency detection and prevention. A time series discord indicates a subsequence that has the maximum difference to the rest of the time series subsequences, meaning that it has abnormal or unusual data trends. In this study, we implemented two versions of time series discord detection algorithms on a high performance parallel database management system (DBMS) and applied them to 240 Hz waveform data collected from 9,723 patients. The initial brute force version of the discord detection algorithm takes each possible subsequence and calculates a distance to the nearest non-self match to find the biggest discords in time series. For the heuristic version of the algorithm, a combination of an array and a trie structure was applied to order time series data for enhancing time efficiency. The study results showed efficient data loading, decoding and discord searches in a large amount of data, benefiting from the time series discord detection algorithm and the architectural characteristics of the parallel DBMS including data compression, data pipe-lining, and task scheduling.

47 OTHER INSTRUMENTATION↗

Detection of Anomalies in Environmental Gamma Radiation Background with Hopfield Artificial Neural Network - Consortium on Nuclear Security Technologies (CONNECT) Q3 Report

Environmental screening of gamma radiation consists of detecting weak nuisance and anomaly signal in the presence of strong and highly varying background. In a typical scenario, a mobile detector-spectrometer continuously measures gamma radiation spectra in short, e.g., one-second, signal acquisition intervals. The measurement data is a 2D matrix, where one dimension is gamma ray energy, and the other dimension is the number of measurements or total time. In principle, gamma radiation sources can be detected and identified from the measured data by their unique spectral lines. Detecting sources from data measured in a search scenario is difficult due to the highly varying background because of naturally occurring radioactive material (NORM), and low signal-to-noise ratio (S/N) of spectral signal measured during one-second acquisition intervals. The objective of this work is to investigate performance of a Hopfield Neural Network (HNN) in in detection and identification of weak nuisances and anomalies events in the presence of a highly fluctuating background. Performance of HNN algorithm is benchmarked using search data from an environmental screening campaign. One data set contained a 137 Cs source, and another dataset contained a 131 I source.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗