With an increasing number of extreme events, grid components and complexity, more alarms are being observed in the power grid control centers. Operators in the control center need to monitor and analyze these alarms to take suitable control actions, if needed, to ensure the system's reliability, stability, security, and resiliency. Although existing alarm and event processing tools help in monitoring and decision making, synchrophasor data along with the topology and component location information can be used in detecting, classifying and locating the event, which is the focus of this work. Phasor Measurement Unit's (PMU's) data quality issue is also addressed before using data for event analysis. The developed algorithms include statistic, clustering, and Maximum Likelihood Criterion (MLE) based anomaly detection, Density-based spatial clustering of applications with noise (DBSCAN) for event detection and physics-based rule/ decision tree for event classification. Further, topology information, statistical techniques, and graph search algorithms are used for event localization. Developed algorithms have been validated with satisfactory results for IEEE 14 bus and 39 Bus as well as with real PMU data from the western US interconnection (WECC).

A Real Time Event Detection, Classification and Localization Using Synchrophasor Data

This paper proposes a novel approach to estimate the fundamental frequency, harmonic and interharmonic phasors in power networks. The proposed method is based on the Matrix Pencil Method (MPM) and Principle Component Analysis (PCA). It estimates parameters of the Prony model that represents the measured signal under noisy conditions. The frequency decomposition feature of the MPM allows concurrent estimation of all frequency components (including all harmonics) of the signal without an additional computational cost. The performance of the proposed method is evaluated with respect to all the requirements defined in the IEEE Std. C37.118 for both M- and P-class phasor measurement units (PMUs). It is shown that the method is compliant with the operating conditions of both PMU classes and at the same time, is able to provide accurate harmonic phasor estimation.

Harmonic and Interharmonic Phasor Estimation Using Matrix Pencil Method for Phasor Measurement Units

Power grid operators assess situational awareness using time-tagged measurements from phasor measurement units (PMUs) placed at multiple locations in a network. However, synchrophasor measurements are prone to anomalies which may impact the performance of phasor based applications. Anomalies include any deviation from expected measurements resulting from power system events or bad data. Bad data include data errors or loss of information due to failures in supporting synchrophasor cyber infrastructure. It is necessary to flag bad data before utilizing for an application. This work proposes a tool for the detection and classification of anomalous data using an unsupervised stacked ensemble learning algorithm. The proposed synchrophasor anomaly detection and classification (SyADC) tool analyzes a selected window of data points using a combination of three unsupervised methods, namely: isolation forest, KMeans and LoOP. The method classifies the data as anomalies or normal data with more than 99% recall. The method also provides a probability of the data to be an event or bad data with more than 99% recall. Results for the IEEE 14 and 68 bus systems with synchrophasor data obtained using Real-Time Digital Simulator and data of industrial PMUs highlight the superiority of the algorithm to detect and classify anomalies.

Real-Time Synchrophasor Data Anomaly Detection and Classification Using Isolation Forest, KMeans, and LoOP

Power systems are moving towards hybrid AC/DC grids with the integration of HVDC links, renewable resources and energy storage modules. The load frequency control (LFC) of tomorrow has to consider the complex interactions between these components. Meanwhile, more attention should be paid to cyber security concerns as the LFC loop highly depends on data communications which may be exposed to cyber attacks. In this regard, this article aims to analyze the false data injection (FDI) attacks on the AC/DC interconnected LFC system with inertia emulation and develop advanced diagnosis tools to reveal their occurrence. We build an optimization-based framework for the purpose of vulnerability analysis. Considering the attack impact on frequency stability, it is shown that the multi-area LFC system with parallel AC/DC links and emulated inertia by storage devices is more vulnerable to FDI attacks, compared to the one without inertia emulation and the normal AC system. We then propose a detection approach to detect and isolate each FDI intrusion with a sufficient fast response, and even recover the attack value. In addition to theoretical results, the effectiveness of the proposed method is validated through simulations on the two-area AC/DC interconnected LFC system with inertia emulation capabilities.

/pdf/false-data-injection-attacks-on-hybrid-ac-hvdc-87d32wmczu.pdf

False Data Injection Attacks on Hybrid AC/HVDC Interconnected Systems With Virtual Inertia—Vulnerability, Impact and Detection

Distributed renewable energy resources (DER) are interfaced in a microgrid through power electronic converters (PEC). PECs utilize different types of sensors for their reliable and stable operation. However, as the ratings of a DER increase, its interfacing PEC becomes more sophisticated requiring more sensors. This will make PECs more vulnerable to sensor malfunctions, which may be caused by cyberattacks. This paper analyzes the performance of a microgrid under sensor malfunction attacks in PECs for different scenarios in MATLAB/SIMULINK and provides a final evaluation based on the acquired results.

Performance Evaluation of an Inverter-Based Microgrid Under Cyberattacks

To guarantee the reliable power supply, the expected operation of all the components in the power system is critical. Distance protection system is primarily responsible of isolating the faulty section from the healthy part of the grid. Failure in protection devices can result in multiple conflicting alarms at the power grid operation center and complex events analysis to manually find the root cause of the observed system state. If not handled in time, it may lead to the propagation of the faults/failures to the adjacent transmission lines and components. With availability of the synchronized measurements from phasor measurement units (PMUs), real-time system monitoring and automated failure diagnosis is feasible. With multiple adverse events and possible data anomalies, the complexity of the problem will be escalated. In this paper, a PMU based algorithm is presented and discussed to detect the root cause of the failure in transmission protection system based on the observed state, e.g. multiple line tripping, breaker failures. The failure diagnosis algorithm is further enhanced to come up with the fully functional version of the failure diagnosis tool, which is tailored for the cases in which the PMU anomalies are present. In the developed algorithm the validity of the PMU data is critical; however, such causes as communication errors or cyber-attacks might lead to the PMU data anomalies. This issue is well-addressed in this paper and some major types of anomaly detection methods suitable for PMU data are discussed. Results show that the ensemble approach has some distinct advantages in data anomaly detection compared to the previously used standalone algorithms. Additionally, the enhanced failure diagnosis method is developed to clean the inaccurate data in case of the anomaly in measured voltage magnitudes. Finally, both original and enhanced versions of the tool are tested on 96-bus test system using the real-time OPAL-RT simulator. The results show the accuracy of the enhanced tool and its advantages over the primary version of the tool.

/pdf/data-driven-failure-diagnosis-in-transmission-protection-sh5e51swai.pdf

Data-driven failure diagnosis in transmission protection system with multiple events and data anomalies

The High Voltage Direct Current (HVDC) technology is typically used for power transfer over longer distances or asynchronous back-to-back connections of the ac networks, or connecting cluster of high power renewable energies to the grid. The performance of the power system can be impacted by the HVDC control and operations. Functionality of the HVDC controllers depend on the information and communication technologies and are vulnerable to cyber-attacks. Hence, the cyber-physical security analysis of the HVDC system needs to be investigated. This study is a step towards cyber-physical analysis for grid with HVDC system by analyzing: 1) the cyber system vulnerabilities of the HVDC system, 2) the integrated cyber-physical system model developement with HVDC technology, 3) the effect of cyber-attacks on the HVDC and impact on the dynamic voltage stability of the power system, and 4) the cyber-attack detection and mitigation. Also, different cyber-attack scenarios on the HVDC system are developed for analyzing the developed approach. The proposed analysis are validated by simulation of the several case studies on the integrated cyber-physical model of the IEEE 9-bus system with the HVDC technology.

Cyber-Physical Vulnerability and Security Analysis of Power Grid with HVDC Line

High penetration of Distributed Energy Resources (DERs), fundamental load behavior changes, controllable loads, and significant increase in Electrical Vehicles (EVs) lead to complex dynamic behavior of the electric distribution system. Increasing number of components also means more measurements, more data and more data anomalies. Detecting, classifying and localizing these anomalies are important for situational awareness, and at the same time, very challenging given increasing complexity of the system. Highly accurate and high-resolution analytical techniques are needed to support anomaly detection, classification and localization (AD-C-L) for monitoring, root cause analysis and decision making. This paper provides comprehensive review and analysis of the existing spatio-temporal AD-C-L techniques within the distribution system. Challenges for specific problems in AD-C- L have been also discussed in this paper. Existing AD-C- L techniques have been categorized and synthesized for specific merits and limitations of multiple Machine Learning (ML) methodologies using common developed metrics of performance. The comparative analysis is summarized and presented with the open research challenges and path forward for future research needs.

Amir Gholami

Papers

Data-driven failure diagnosis in transmission protection system with multiple events and data anomalies

Cyber-Physical Vulnerability and Security Analysis of Power Grid with HVDC Line

Comparative Analysis of ML Techniques for Data-Driven Anomaly Detection, Classification and Localization in Distribution System