Search NASA⌕ Search

Engineering topics

Yan, Weizhong

Publications and source records attributed to Yan, Weizhong.

Attack detection and localization with adaptive thresholding

According to some embodiments, a system, method, and non-transitory computer readable medium are provided comprising a plurality of real-time monitoring nodes to receive streams of monitoring node signal values over time that represent a current operation of the cyber physical system; and a threat detection computer platform, coupled to the plurality of real-time monitoring nodes, to: receive the monitoring node signal values; compute an anomaly score; compare the anomaly score with an adaptive threshold; and detect that one of a particular monitoring node and a system is outside a decision boundary based on the comparison, and classify that particular monitoring node or system as anomalous. Numerous other aspects are provided.

Abbaszadeh, Masoud↗

Systems and methods for global cyber-attack or fault detection model

An industrial asset may have monitoring nodes that generate current monitoring node values representing a current operation of the industrial asset. An abnormality detection computer may detect when a monitoring node is currently being attacked or experiencing a fault based on a current feature vector, calculated in accordance with current monitoring node values, and a detection model that includes a decision boundary. A model updater (e.g., a continuous learning model updater) may determine an update time-frame (e.g., short-term, mid-term, long-term, etc.) associated with the system based on trigger occurrence detection (e.g., associated with a time-based trigger, a performance-based trigger, an event-based trigger, etc.). The model updater may then update the detection model in accordance with the determined update time-frame (and, in some embodiments, continuous learning).

Xu, Rui↗

Systems and methods for enhanced sequential power system model parameter estimation

A system for enhanced sequential power system model calibration is provided. The system is programmed to store a model of a device. The model includes a plurality of parameters. The system is also programmed to receive a plurality of events associated with the device, receive a first set of calibration values for the plurality of parameters, generate a plurality of sets of calibration values for the plurality of parameters, for each of the plurality of sets of calibration values, analyze a first event of the plurality of events using a corresponding set of calibration values to generate a plurality of updated sets of calibration values, analyze the plurality of updated sets of calibration values to determine a current updated set of calibration values, and update the model to include the current updated set of calibration values.

Wang, Honggang↗

Cyber-Physical Resiliency for Wind Power Generation

During the next three years, wind power generation is expected to add more generation capacity to the national electrical grid than any other energy sector. With new wind turbine designs rated power over 10 MW and wind farms reaching over 1 GW capacities, the consequences of cyber-attacks on wind power generation are becoming increasingly more critical. Moreover, the known vulnerabilities of wind turbine control systems and the potential damaging effects of intrusions, motivate an urgent protection improvement for the wind power generation sector. This program has developed a variety of new adaptive defense technologies that enable wind power generation systems to survive sophisticated cyberattacks by enhancing the control systems capabilities of detection, localization, and accommodation. The introduction of these technologies in the on-shore and the emerging off-shore market will result in a significantly more reliable and secure wind power infrastructure.

17 WIND ENERGY↗

Multivariate Time Series Anomaly Detection with Few Positive Samples

Given the scarcity of anomalies in real-world applications, the majority of literature has been focusing on modeling normality. The learned representations enable anomaly detection as the normality model is trained to capture certain key underlying data regularities under normal circumstances. In practical settings, particularly industrial time series anomaly detection, we often encounter situations where a large amount of normal operation data is available along with a small number of anomaly events collected over time. This practical situation calls for methodologies to leverage these small number of anomaly events to create a better anomaly detector. In this paper, we introduce two methodologies to address the needs of this practical situation and compared them with recently developed state of the art techniques. Our proposed methods anchor on representative learning of normal operation with autoregressive (AR) model along with loss components to encourage representations that separate normal versus few positive examples. We applied the proposed methods to two industrial anomaly detection datasets and demonstrated effective performance in comparison with approaches from literature. Our study also points out additional challenges with adopting such methods in practical applications.

Xue, Feng↗

Cyber Physical Protection for Natural Gas Compression

The secure transport of natural gas through North American pipelines is vital for both home and industrial purposes. GE in collaboration with Baker Hughes and Idaho National Laboratory, completed a US Department of Energy grant program to develop a new cyber-physical protection system that monitors the large compressor stations used to boost pressure and maintain proper gas flow. Algorithms were developed to detect the presence of a cyber-attack that impacts the compressor station physical processes, locate the critical functions being manipulated, and potentially neutralize the attack allowing continued operations. The technology was successfully demonstrated at an operating compressor facility located in New York state.

03 NATURAL GAS↗

Cyber-Attack Detection and Accommodation for the Energy Delivery System

The goals of this project were to create a software system with a suite of key algorithms for cyber-attack detection and accommodation providing domain layer protection for critical power generation assets. Example assets included gas and steam turbines, heat recovery steam generators, and electrical generators. The aggressive algorithm goals were aimed at reducing the false positive rates in threat detection to <1% using learnings from many evolving disciplines (power turbine and generator physics, power system modeling, modern control theory, system identification, machine learning, deep learning, mathematics and data science). Additional goals for the algorithms involved localizing threats on-the-fly to know in which monitoring node the effects of attacks are present, and then providing accommodation to keep the system running uninterrupted much of the time in the presence of the attack. Accommodation had a performance goal of providing resiliency when up to 50% of monitoring nodes are in an attack state.

cybersecurity, cyber-physical↗

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↗

Deep Analysis Net with Causal Embedding for Coal-fired Power Plant Fault Detection and Diagnosis (DANCE4CFDD)

Fault detection and diagnosis is critical to power plant operation to ensure attaining high reliability while reducing operation cost. As more renewable power is introduced to the power grid, traditional fossil power plants take on the extra burden of excessive load cycling to compensate the generation variability from renewable power. Such load cycling will pose more reliability challenges to power plant operation. There are a number of challenges faced by today’s asset health management system in coal- fired (or gas) power plants: 1) high-dimensional nonlinear interaction among multiple time series measurements; 2) high measurement variance induced by operational conditions/modes; 3) variation among asset types and plant configurations; and 4) a small number of faulty events to learn from. To cope with these challenges, today’s fielded asset health management systems rely heavily on manual efforts from domain experts and hand-crafted features or rules based on domain knowledge. Despite its role in plant reliability, such a practice is costly and hinders its scalability and sustainability, particularly when a plant undergoes modifications. The objective of this project is to develop a novel end-to-end AI learning system that is trainable (i.e., the AI representation of a complex system behavior can be directly learned from properly labeled data) for accurate fault detection and root cause analysis. The ability to create a fault detection model directly from time series could alleviate the efforts associated with today’s asset management solution development. In the course of this project, we have achieved the following: Created an AI model development environment incorporating state-of-the-art neural network architectures for rapid model development and evaluation; Developed novel learning strategies for training of fault detection model; Developed special-purpose neural network architecture embedded with variable association graph aiming for better interpretability; Developed a learning strategy to leverage a small number of faulty events for enhanced fault detection capability; Conducted detailed experimental study based on public benchmark datasets and demonstrated the effectiveness of the proposed solution; and Validated the developed system with data from both a coal-fired plant boiler dynamic simulation model and real-world coal-fired power plant covering multiple asset and fault types. Overall, the project attained a technology readiness level of TRL 5 from TRL 2 at the beginning of the project.

20 FOSSIL-FUELED POWER PLANTS↗

Local and global decision fusion for cyber-physical system abnormality detection

Monitoring nodes may generate a series of current monitoring node values over time representing current operation of a cyber-physical system. A decision fusion computer platform may receive, from a local status determination module, an indication of whether each node has an initial local status of “normal”/“abnormal” and a local certainty score (with higher values of the local certainty score representing greater likelihood of abnormality). The computer platform may also receive, from a global status determination module, an indication of whether the system has an initial global status of “normal”/“abnormal” and a global certainty score. The computer platform may output, for each node, a fused local status of “normal” or “abnormal,” at least one fused local status being based on the initial global status. The decision fusion computer platform may also output a fused global status of “normal” or “abnormal” based on at least one initial local status.

Abbaszadeh, Masoud↗

Big Data Analysis of Massive PMU Datasets: A Data Platform Perspective

The discovery of `event signatures' and useful insights from very large historical Phasor Measurement Unit (PMU) datasets is predicated on offline Big Data analysis approaches that rely on the generation of predictive features on a massive scale. This paper presents lessons learned from a data platform perspective towards reducing barriers to adoption of Big Data analytics against a real dataset of almost half a trillion data points drawn from over 400 PMUs distributed across the North American power grid. We demonstrate software abstractions and targeted performance optimizations that can lead to significant productivity gains for power systems researchers seeking to perform offline exploratory temporal analysis and modeling tasks, with a focus on feature generation. We describe how our optimized approach goes beyond a naive application of mainstream Big Data technologies, enabling feature generation tasks, that previously took days or even weeks, to now be completed in just a few hours.

Kumar, Vijay S.↗

Deep anomaly detection for industrial systems: a case study

We explore the use of deep neural networks for anomaly detection of industrial systems where the data are multivariate time series measurements. We formulate the problem as a self-supervised learning where data under normal operation are used to train a deep neural network autoregressive model, i.e., use a window of time series data to predict future data values. The aim of such a model is to learn to represent the system dynamic behavior under normal conditions, while expect higher model vs. measurement discrepancies under faulty conditions. In real world applications, many control settings are discrete in nature. In this paper, vector embedding and joint losses are employed to deal with such situations. Both LSTM and CNN based deep neural network backbones are studied on the Secure Water Treatment (SWaT) testbed datasets. Also, Support Vector Data Description (SVDD) method is adapted to such anomaly detection settings with deep neural networks. Evaluation methods and results are discussed based on the SWaT dataset along with potential pitfalls.

anomaly detection, deep neural network↗

Multi-class decision system for categorizing industrial asset attack and fault types

According to some embodiments, a plurality of monitoring nodes may each generate a series of current monitoring node values over time that represent a current operation of the industrial asset. A node classifier computer, coupled to the plurality of monitoring nodes, may receive the series of current monitoring node values and generate a set of current feature vectors. The node classifier computer may also access at least one multi-class classifier model having at least one decision boundary. The at least one multi-class classifier model may be executed and the system may transmit a classification result based on the set of current feature vectors and the at least one decision boundary. The classification result may indicate, for example, whether a monitoring node status is normal, attacked, or faulty.

97 MATHEMATICS AND COMPUTING↗

Forecasting and Early Warning for Adversarial Targeting in Industrial Control Systems

A new feature-based forecasting framework of abnormalities is presented for early warning for cyber-physical control systems. The abnormalities may refer to intelligent cyber-attacks or naturally occurring faults and failures. Techniques presented here are aimed at protecting against unauthorized intrusions as well as fault prevention. Time series signals from system monitoring nodes are converted to features using feature discovery techniques. The feature behavior for each monitoring node is characterized in the form of decision boundaries, separating normal and abnormal space with operating data collected from the plant or by running virtual models of the plant. A set of ensemble state-space models are constructed for representing feature evolution in time-domain, where the ensembles are selected using Gaussian Mixture Model (GMM) clustering. The forecasted outputs are anticipated time-evolution of features, computed by applying an adaptive Kalman predictor to each ensemble model. The overall features forecast is then obtained through dynamic ensemble averaging. This is done by projecting evolution of feature vector to future times in a receding horizon fashion. The features forecast are compared to the decision boundary to estimate if/when the feature vectors will cross the boundary. Simulation results in a high fidelity GE gas turbine platform, show the effectiveness of our approach for forecasting abnormalities.

42 ENGINEERING↗

Cyberattack Detection for Cyber Physical Systems Security–A Preliminary Study.

Cyber-physical systems (CPS) security has become an increasingly important research topic in recent years. Geared towards more advanced cyberattack detection techniques as part of strategies for enhancing the security of CPS, in this paper we propose a machine learning based cyber-attack detection scheme. The proposed scheme is a physical-domain technique; specifically, it assumes the physical measurements of the system carry sufficient information for capturing the system behavior, thus can be used for differentiating normal operation and attacks. CPS are complex in nature and the number of physical measurements available for CPS is often overwhelmingly high. Thus, accurately modeling CPS’ dynamic behavior, more importantly, distinguishing normal and adversary activities based on the large number of physical measurements, can be challenging. To address the challenge, we have focused our research effort on feature engineering, that is, to intelligently derive a set of salient signatures or features from the noisy measurements. We make sure the derived features are more compact and, more importantly, have more discriminant power than the original physical measurements, thus enabling us to achieve more accurate and robust detection performance. To demonstrate the effectiveness of the proposed scheme, in our experimental study we consider gas turbines of combined cycle power plants as the cyberphysical system. Using the data from the high-fidelity simulation we show that our proposed cyberattack detection scheme is able to achieve high detection performance.

97 MATHEMATICS AND COMPUTING↗