Search NASA⌕ Search

SEARCH · Search NASA

Results for “Anomaly location and classification”

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.

Multi-Source Data Aggregation and Real-Time Anomaly Classification and Localization in Power Distribution Systems

This paper proposes a real-time anomaly location and classification framework for power distribution systems to simultaneously determine the type of anomaly (i.e., short-circuit fault, cyber attack, DER switching) and its location. The proposed framework employs the data aggregation module to collect the measurement data from multiple field devices operating at different sampling rates, such as protection relays and D-PMUs. The output of the data aggregation is then fed into a multi-task learning-based long-based short-term memory (MTL-LSTM) to classify the type of anomaly and the location in two separate tasks. The proposed MTL-LSTM approach can be utilized in real-time operation in order to distinguish between normal and several anomalous operations and locate the anomaly. The proposed framework is tested on a modified IEEE 33-bus test feeder benchmark that integrates solar generation and energy storage. Furthermore, the results show that the proposed framework can locate and classify anomalies for several operation conditions with more than 96% accuracy. Further experiments highlight the impact of aggregating multiple sources of data on the performance of the proposed model.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Performance Evaluation of Vertical Federated Machine Learning Against Adversarial Threats on Wide-Area Control System: Preprint

Federated machine learning (FL) is gaining significant popularity to develop cybersecurity solutions in power grids because of its advanced capability to support decentralized data handing at local devices, its privacy preservation, and its low-bandwidth requirement. However, the evolving adversarial machine learning (AML) threats raise significant concerns for the cybersecurity of FL architectures. The FL-based split neural network (SplitNN) achieves high performance through the decentralized training of local neural network models while preserving data privacy across multiple entities. In this paper, we propose a methodology for evaluating the performance of a vertical FLbased anomaly detector against different types of AML attacks, including denial-of-service attacks, adversarial data injection attacks, and replay attacks on the trained local models deployed in the grid network. For a case study, we consider the modified IEEE 13-bus system, and we develop SplitNN-based binary and multiclass classification models to detect, locate, and identify different types of data integrity attacks on the volt-watt control with two pooling layers: maximum pooling and AvgPool. Our experimental results, computed through performance metrics, reveal that the severity of these AML attacks varies with the integrated pooling mechanism, the type of classification model, and the nature of the cyberattack. Further, the AML attacks negatively impacted the prediction time per sample for the pretrained SplitNN during the online testing.

adversarial threats↗

Primer on Use of Multi-Spectral and Infra Red Imaging for On-Site Inspections

The purpose of an On-Site Inspection (OSI) is to determine whether a nuclear explosion has occurred in violation of the Comprehensive Nuclear Test Ban Treaty (CTBT), and to gather information which might assist in identifying the violator (CTBT, Article IV, Paragraph 35) Multi-Spectral and Infra Red Imaging (MSIR) is allowed by the treaty to detect observables which might help reduce the search area and thus expedite an OSI and make it more effective. MSIR is permitted from airborne measurements, and at and below the surface to search for anomalies and artifacts (CTBT, Protocol, Part II, Paragraph 69b). The three broad types of anomalies and artifacts MSIR is expected to be capable of observing are surface disturbances (disturbed earth, plant stress or anomalous surface materials), human artifacts (man-made roads, buildings and features), and thermal anomalies. The purpose of this Primer is to provide technical information on MSIR relevant to its use for OSI. It is expected that this information may be used for general background information, to inform decisions about the selection and testing of MSIR equipment, to develop operational guidance for MSIR use during an OSI, and to support the development of a training program for OSI Inspectors. References are provided so readers can pursue a topic in more detail than the summary information provided here. The following chapters will provide more information on how MSIR can support an OSI (Section 2), a short summary what Multi-Spectral Imaging and Infra Red Imaging is (Section 3), guidance from the CTBT regarding the use of MSIR (Section 4), and a description of several nuclear explosion scenarios (Section 5) and consequent observables (Section 6). The remaining sections focus on practical aspects of using MSIR for an OSI, such as specification and selection of MSIR equipment, operational considerations for deployment of MISR equipment from an aircraft, and the conduct of field exercises to mature MSIR for an OSI. Finally, an appendix provides detail describing the magnitude and spatial extent of the surface shock expected from an underground nuclear explosion. If there is a seismic event or other data to suggest there has been a nuclear explosion in violation of the CTBT, an OSI may be conducted to determine whether a nuclear explosion has occurred and to gather information which may be useful in identifying the party responsible for conducting the explosion. The OSI must be conducted in the area where the event that triggered the inspection request occurred, and the inspected area must not exceed 1,000 square kilometers, or be more than 50 km on aside (CTBT Protocol, Part II, Paragraphs 2 and 3). One of the guiding principles for an inspection is that it be effective, minimally intrusive, timely, and cost-effective [Hawkins, Feb 1998]. In that context, MSIR is one of several technologies that can be used during an aircraft overflight to identify ground regions of high interest in a timely and cost-effective manner. This allows for an optimized inspection on the ground. The primary purpose for MSIR is to identify artifacts and anomalies that might be associated with a nuclear explosion, and to use the location of those artifacts and anomalies to reduce the search area that must be inspected from the ground. The MSIR measurements can have additional utility. The multi-spectral measurements of the ground can be used for terrain classification, which can aid in geological characterization of the Inspected Area. In conditions of where light smoke or haze is present, long-wave infrared imaging can provide better imaging of the ground than is possible with standard visible imagery.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗

MODAL CLASSIFICATION: Processing and Anomaly Detection in a Noisy Vibrational Environment

n this report we summarize the development of a model-based detection scheme capable of the on-line detection of structural anomalies applying both system identification methods to extract a modal model and state estimation/detection methods to both detect an anomaly and locate all affected sensor positions specified by measurement location. Background information evolving from a simple single input/single output (SISO) mechanical system introducing the state-space formulation which leads directly to a multiple input/multiple output (MIMO) state-space formulation is discussed.From that representation the modal state-space model is introduced through an eigenvalue-eigenvector formulation leading to a similarity transformation and complex modes. Next the sampled-data state-space system is introduced leading to discrete state-space systems which provide the basis for the system identification methods to follow.

42 ENGINEERING↗

The Retrograde Transvenous Push-Through Method: A Novel Treatment of Peripheral Arteriovenous Malformations with Dominant Venous Outflow

PurposeTo evaluate the efficacy and safety of a novel retrograde transvenous embolization technique of peripheral arteriovenous malformations (AVMs) using Onyx.Materials and MethodsWe conducted a retrospective analysis of all patients who underwent transvenous retrograde Onyx embolization of peripheral AVMs with dominant venous outflow over a 29-month period. The embolization is aimed at retrograde filling of the nidus after building a solid plug in the dominant venous outflow (push-through). Classification, clinical signs, technical aspects, clinical and technical success rates, and complications were recorded. Short-term outcome was assessed.Results11 Symptomatic patients (8 female; mean age 31.4 years) were treated at our Vascular Anomalies Center with this method between January 2012 and May 2014. The AVMs were located on the upper extremity (n = 3), pelvis (n = 2), buttock (n = 2), and lower extremity (n = 4). Retrograde embolization was successfully carried out after preparatory transarterial-flow reduction in eight cases (73 %) and venous-flow reduction with Amplatzer Vascular Plugs in four cases (36 %). Complete devascularization (n = 10; 91 %) or 95 % devascularization (n = 1; 9 %) led to complete resolution (n = 8; 73 %) or improvement of clinical symptoms (n = 3; 27 %). One minor complication occurred (pain and swelling). During a mean follow-up time of 8 months, one clinically asymptomatic recurrence of AVM was detected.ConclusionInitial results suggest that retrograde transvenous Onyx embolization of peripheral AVMs with dominant venous outflow is a safe and effective novel technique with a low complication rate.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Phasor-Measurement-Unit-Based Data Analytics Using Digital Twin and PhasorAnalytics Software

A major objective of this project was to apply GE’s commercial machine learning and data analytics toolsets to large-scale, real-world, anonymized Phasor Measurement Unit (PMU) datasets in order to extract signatures, correlated and/or causal factors, and precursor patterns associated with significant power system phenomena. The project had a particular emphasis on extraction of insights relevant to asset health monitoring, real-time load modeling and cybersecurity monitoring. Additionally, the team was directed to undertake a comprehensive data quality analysis for the provided datasets and encouraged to estimate the ‘machine-learning readiness’ of the datasets by documenting any major obstacles to the application of commercial machine learning algorithms. To accomplish the aforementioned objectives, the project team’s work centered around the identification of key event signatures and application of the identified event signatures for event detection and event classification. The industry-validated, semi-supervised machine learning strategy employed for event signature identification involved several major tasks, including data-preprocessing, generation of an overabundance of features, normal data identification, normality modeling, and event signature identification through a methodical, quantitative ranking of features in order of relevance to each studied event type. Throughout the project, data quality issues and mitigation techniques were investigated. In this report, insights are provided regarding the readiness of the provided synchrophasor datasets for application of machine learning and data analytics. The methodologies employed for this technical strategy are summarized in this report. With regards to data preprocessing and feature generation, the provided Training and Test Datasets were ingested into GE’s big data environment. Subsequently, the team applied bad data cleansing and data imputation scripts, event detection scripts, and application programming interfaces (APIs) to the datasets for convenient data access. The project team completed development and validation of dozens of physics-based, statistics-based and transformation-based feature functions used for the extraction of over 60 synchrophasor features. Using a new parallel feature generation technology developed on this project, over 60 features have been rapidly generated for the full two years’ worth of Training and Test Dataset data associated with both the Eastern and Western interconnects. Even accommodating for temporal down-sampling inherent to the feature extraction procedure, this parallel feature generation activity resulted in a massive feature set with a storage requirement approximately equal to that of the raw training dataset itself. With regards to normal data identification and normality modeling, a normality model was built using the feature data extracted from the Training Dataset and iteratively refined subsequent to incremental adjustments and expansions of the Training Dataset feature data. With respect to event characterization and signature identification, an event signature identification pipeline was developed and used in conjunction with the normality model to identify over 15 event signatures for key event categories within the Training Dataset. The identified event signatures were used to characterize hundreds of key events in terms of relative severity, duration, and location of the event. An investigation was undertaken to identify correlated and causal factors involved in transformer events. A separate investigation into temporal trends in ring-down analysis results was undertaken to determine possible associations between system dynamics and various other factors such as loading, season or year. To validate the identified event signatures, additional work was undertaken to develop signature-based anomaly detection and classification tools suitable for convenient application to the synchrophasor datasets. The anomaly detection and classification tools, suitable for online application, were then applied to the entirety of the Eastern Interconnect Training and Test Datasets. Performance of the event detection and classification tools was evaluated upon receipt of the Test Dataset event logs (i.e., the labels for events contained in the Test Dataset), and promising results were obtained despite several challenges (documented herein) associated with application of supervised or semi-supervised machine learning methods to large-scale, anonymized datasets. Finally, the detection and classification tools were used to detect, classify, and characterize thousands of new events not included in the original event logs provided by the DOE within both the Training and Test Datasets.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Detection of Synchrophasor False Data Injection Attack using Feature Interactive Network

The synchrophasor data recorded by Phasor Measurement Units (PMUs) plays an increasingly critical role in the regulation and situational awareness of power systems. However, the widely installed PMUs are vulnerable to multiple malicious attacks from cyber hackers during data transmission and storage. To address this problem, a Modified Ensemble Empirical Mode Decomposition (MEEMD) is proposed first to extract the intrinsic mode functions of each Synchrophasor Data Attacks (SDA). The frequency-based adaptive screening criterion embedded in MEEMD is used to eliminate the false intrinsic mode functions. Next, a Multivariate Convolutional Neural Network (MCNN) is proposed to identify multiple SDA by utilizing the extracted intrinsic mode functions and original SDA as input vectors. A fusion block as the main structure of MCNN is also leveraged to increase the diversity of features and compress the model parameters. Integrating MEEMD and MCNN, a framework with automatic feature extraction and multi-source information fusion capability, referred to as Feature Interactive Network (FIN), is proposed to detect multiple SDA. Based on the proposed FIN framework, six types of SDA are explored for the first time using actual synchrophasor data in FNET/Grideye that was collected from different locations in the U.S. Eastern Interconnection. Finally, a large quantity of experiments with different attack strengths are used to evaluate the adaptability and classification performance of the proposed FIN.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Regime Characterization of Offshore Wind Resource Using Unsupervised Learning

Predictability of wind resource conditions is critical for offshore wind design and operations. While many studies of extreme wind conditions focus on specific events such as low-level jets or ramps, these rely on threshold definitions that limit generality. Here we present a data-driven framework that combines principal component analysis (PCA), self-organizing maps (SOM), and k-means clustering to classify wind resource conditions as typical and anomalous from climatological data. Anomalies are defined not by fixed thresholds but by flagging samples located far from SOM node centers inside the baseline SOM structure. This reframes extremes as rare ebents and hence, likely difficult to anticipate by numerical weather prediction models. We applied this approach to 23 years (2000–2022) of hourly profiles from the NOW-23 hindcast model at the Humboldt Wind Energy Area. Classification is conducted on a feature space consisting of 10 m wind speed and direction, bulk shear and veer across 30–270 m, and a low-level jet index. Dimensionality reduction is achieved through PC. A 2 × 3 OM lattice trained on the PCA vectors identified six baseline regimes spanning weak to strong flow states. High quantization-error profiles are identified and re-clustered into four anomalous regimes. The baseline regimes exhibited clear seasonal and diurnal cycles. Meanwhile, the anomalous regimes represented <10 % of all hours but showed distinct combinations of speed, shear, and veer, when compared to the baseline regimes. Anomalous regimes are typically short-lived (~few hours), yet their transitions can lead to hub-height wind changes of −18 to +9 m s -1 . For a representative 15 MW turbine, these shifts imply rapid swings in capacity factor from near-full output to negligible generation. Validation with lidar buoy data showed 51% agreement in SOM labels across ~6,000 overlapping hours, with most mismatches confined to adjacent speed classes. HRRR comparisons further revealed that anomalous regimes were disproportionately associated with forecast biases exceeding 5 m s -1 . Together, these results reframe extremes in offshore wind from absolute maxima or minima to weather states that are difficult to anticipate from models.

17 WIND ENERGY↗

Characterization of Used Nuclear Fuel with Multivariate Analysis for Process Monitoring

The Multi-Isotope Process (MIP) Monitor combines gamma spectroscopy and multivariate analysis to detect anomalies in various process streams in a nuclear fuel reprocessing system. Measured spectra are compared to models of nominal behavior at each measurement location to detect unexpected changes in system behavior. In order to improve the accuracy and specificity of process monitoring, fuel characterization may be used to more accurately train subsequent models in a full analysis scheme. This paper presents initial development of a reactor-type classifier that is used to select a reactor-specific partial least squares model to predict fuel burnup. Nuclide activities for prototypic used fuel samples were generated in ORIGEN-ARP and used to investigate techniques to characterize used nuclear fuel in terms of reactor type (pressurized or boiling water reactor) and burnup. A variety of reactor type classification algorithms, including k-nearest neighbors, linear and quadratic discriminant analyses, and support vector machines, were evaluated to differentiate used fuel from pressurized and boiling water reactors. Then, reactor type-specific partial least squares models were developed to predict the burnup of the fuel. Using these reactor type-specific models instead of a model trained for all light water reactors improved the accuracy of burnup predictions. The developed classification and prediction models were combined and applied to a large dataset that included eight fuel assembly designs, two of which were not used in training the models, and spanned the range of the initial 235U enrichment, cooling time, and burnup values expected of future commercial used fuel for reprocessing. Error rates were consistent across the range of considered enrichment, cooling time, and burnup values. Average absolute relative errors in burnup predictions for validation data both within and outside the training space were 0.0574% and 0.0597%, respectively. The errors seen in this work are artificially low, because the models were trained, optimized, and tested on simulated, noise-free data. However, these results indicate that the developed models may generalize well to new data and that the proposed approach constitutes a viable first step in developing a fuel characterization algorithm based on gamma spectra.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗