Real-Time Disturbance Detection and Classification using Principal Component Analysis of PMU Data

This paper concerns with the production of synthetic phasor measurement unit (PMU) data for research and education purposes. Due to the confidentiality of real PMU data and no public access to the real power systems infrastructure information, the lack of credible realistic data becomes a growing concern. Instead of constructing synthetic power grids and then producing synthetic PMU measurement data by time simulations, we propose a model-free approach to directly generate synthetic PMU data. we train the generative adversarial network (GAN) with real PMU data, which can be used to generate synthetic PMU data capturing the system dynamic behaviors. To validate the sequential generation by GAN to mimic PMU data, we theoretically analyze GAN's capacity of learning system dynamics. Further by evaluating the synthetic PMU data by a proposed quantitative method, we verify GAN's potential to synthesize realistic samples and meanwhile realize that GAN model in this paper still has room to improve. Moreover it is the first time that such generative model is applied to synthesize PMU data.

This study proposes a deep learning-based fault detection method for intelligent distribution networks using Phasor Measurement Unit (PMU) data. With the increasing development of intelligent distribution systems, the need for fast and accurate fault detection systems is crucial to improve the reliability and resilience of the power grid. Utilizing PMU data, which provides real-time information on voltage, current, and frequency, enables more precise and rapid fault detection. In this study, we developed a deep learning model that uses Long Short-Term Memory (LSTM) to sequentially process PMU data and detect faults such as short circuits, phase faults, and line outages. The model was trained on a PMU dataset covering a wide range of normal and fault conditions in the distribution network. Evaluation results show that the proposed model is capable of detecting faults with >98% accuracy and has a faster detection time compared to traditional detection methods. This approach also demonstrates the ability to identify fault types with a high degree of reliability and reduces the risk of system failure due to detection delays. By using deep learning methods, this study contributes to improving the reliability of intelligent distribution systems and provides a basis for the application of PMU technology in more efficient and automated distribution network monitoring and maintenance.

Synchronized phasor measurement unit (PMU) data contain rich information about power systems, hence the quality of estimates is of utmost concern. The existing methodologies for estimation rely on certain assumptions regarding the error in the measurement data. This paper revisits a key assumption specifically the Gaussian character of the error. A quantification of the PMU error yields its nature and statistical properties including its dependence on various sections of the PMU instrumentation channel (supposedly the major source of error in the PMU data). The non Gaussian nature of the error is asserted using various null hypotheses tests and a novel Gaussian mixture model based clustering technique is proposed to characterize and relate the errors present in PMU measurement data to the saturation in current transformer, cable length and the PMU burden. The proposed approach is tested using both real and synthetic PMU datasets. The ultimate goal of the paper is towards creating a PMU error emulator for testing and research of data analytic algorithms focused on crucial WAMS based applications.

Introduction of Synchronized phasor measurement unit (PMU) in power system is a solution for monitoring system wide disturbances. PMUs provide precise information about a system’s phasors such as magnitude and phase angle of sine waves of the system. PMU data is a valuable source when determining the post-mortem of a fault, or a system disturbance. It can be unveiled what time the system began to falter and exactly what instabilities the system experienced. Knowing the cause of such instabilities will allow enhancement of the current system protection system. This will help in minimizing risk, disruptions or total system collapse. PMUs are synchronized via global positioning system (GPS). For wide area monitoring (WAM), synchronization of the PMUs allows for interconnected systems to be monitored simultaneously, giving real time records. This paper explores the impact of PMU in modern power system, deployment strategies of PMU network around Texas Tech University, importance of PMU data collected from the network, correlation of events by data analysis and focus on how the time stamped information is valuable for grid stability. The ongoing smart architecture of grid will increase dependency and importance on PMU based network in future.

This paper studies the fundamental dimensionality of synchrophasor data, and proposes an online application for early event detection using the reduced dimensionality. First, the dimensionality of the phasor measurement unit (PMU) data under both normal and abnormal conditions is analyzed. This suggests an extremely low underlying dimensionality despite the large number of the raw measurements. An early event detection algorithm based on the change of core subspaces of the PMU data at the occurrence of an event is proposed. Theoretical justification for the algorithm is provided using linear dynamical system theory. Numerical simulations using both synthetic and realistic PMU data are conducted to validate the proposed algorithm.

A PMU data recovering method based on preferred selection strategy

This paper presents a multiscale phasor measurement unit (PMU) data-compression method based on clustering analysis of wide-area power systems. PMU data collected from wide-area power systems involve local characteristics that are significant risk factors when applying dimensionality-reduction-based data compression. Therefore, density-based spatial clustering of applications with noise (DBSCAN) is proposed for the preconditioning of PMU data, except for bad data and the automatic segmentation of correlated local datasets. Clustered PMU datasets of a local area are then compressed using multiscale principal component analysis (MSPCA). When applying MSPCA, each PMU signal is decomposed into frequency sub-bands using wavelet decomposition, approximation matrix, and detail matrices. The detail matrices in high-frequency sub-bands are compressed by using a PCA-based linear-dimensionality reduction process. The effectiveness of DBSCAN for data compression is verified by application of the proposed technique to the real-world PMU voltage and frequency data. In addition, comparisons are made with existing compression techniques in wide-area power systems.

Chapter 6 - Application of the phasor measurement unit for protecting unintentional islanding of the distribution system

An end-to-end supervised learning method is proposed for fault detection in the electric grid using Big Data from multiple Phasor Measurement Units (PMUs). The approach consists of preprocessing steps aimed at reducing data noise and dimensionality, followed by utilization of six classification models considered for detecting faults. Three of the models were variants of Convolutional Neural Network (CNN) architectures that consider a single type of measurement (voltage, current or frequency) at all PMUs or all types together also at all PMUs. CNN based models were compared to traditional methods of Logistic Regression (LR), Multi-layer Perceptron (MLP) and Support Vector Machine (SVM). Evaluation was conducted on two-year data measured by PMUs at 37 locations in a large electric grid. The response variable for classification were extracted from the grid-wide outage event log. Experiments show that CNN-based models outperformed traditional methods on one year out-of-sample outage detection over the entire grid.

The widespread implementation of Phasor Measurement Units (PMUs) on power systems has facilitated the development of accurate and fast dynamic state estimators (DSE), which helps to reliably achieve the real-time monitoring and control of the power systems dynamics. Dynamic state estimators are used by the investigators to estimate and identify the state variables and other parameters associated with the nonlinear dynamic models within the power systems, by using the measurements which are mainly provided by the PMUs. This paper primarily tries to present an innovative and robust dynamic state estimation approach when the PMU data are missing due to various reasons. The highly efficient Unscented Kalman Filter (UKF) algorithm is used to formulate the DSE. The simulation results on IEEE 14 bus system shows that the proposed algorithm remains robust and produce accurate estimates during the missing of PMU data coming from generator buses.

The excitation system is an important system of power plants, which has an effective role in power system dynamic and stability. Consequently, developing an approach for the excitation system's model and parameter estimation is a necessary research goal. In real situations, it is often needed to identify and estimate unknown parameters of the excitation system by field recorded signals. The recorded signal can be internal from the distributed control system (DCS) or external from the phasor measurement unit (PMU). In this article, a novel method based on cubature Kalman filter and data mining is proposed to identify the parameters of a standard type excitation system. Firstly, the best available signals for parameters estimation are chosen. Secondly, a method is proposed to parameters estimation of excitation systems efficiently, when both the DCS and PMU with different sample rates are employed to record the measurement data. Data mining is performed at intervals that DCS data is missing (or unavailable), while PMU data is available. Experimental data are used for validation of the proposed approach.

We introduce a new method for online Phasor Measurement Unit (PMU) missing value replacement. Our approach allows us to decompose PMU event responses into a non-dynamic component (denoted the participation factor) that can be inferred directly from the past and a dynamic component that can be inferred directly from all other PMUs (denoted the event strength). When missing values occur, we can use these two components, which do not rely on the missing index, to estimate the correct value. The method is extremely fast and can easily be used for online applications. Furthermore, extensive testing on real power system event data reveals that our approach achieves state-of-the-art performance in terms of Mean Absolute Percent Errors (MAPEs) for PMU data dropped during event periods. The method also yields an interpretable and simplified view of events for further analysis and applications. The method relies only on PMU data and does not take outside information such as network topology.

The characteristics of oscillation modes, such as interarea, regional, and subsynchronous modes, can vary during a power system fault, which can cause switching and control actions in the power system. Transient data of the modal response due to such a fault can be acquired through phasor measurement units (PMUs). When the transient data have a long duration, it is desirable to perform modal identification separately on each segment of the transient data, so as to reflect the varying characteristics of oscillation modes. A conventional discrete Fourier transform (DFT)-based method for parametric modal identification cannot be efficiently applied to such a segment dataset. In this paper, a DFT-based method with an exponential window function is proposed to identify oscillation modes from each segment of transient data in PMUs. This window function allows the application of the least squares method (LSM) for modal identification in the same manner as the conventional method. The accuracy of identification of the proposed method is compared with those of the conventional method and a Prony method through synthetic data of transient responses. Its feasibility is also verified by identifying real-world oscillation modes from transient data in PMUs.

Large databases and data warehouses are becoming prevalent for the storage and management of energy data. Accelerating the rates at which data can be retrieved is beneficial not only to allow for more efficient search of the data, but also to be integrated with other energy system tools. In this paper, a fast indexing and data retrieval method, known commonly as a bitmap index, is created to facilitate intelligent querying and indexing of data generated by phasor measurement units (PMU) at a rate of 60 Hz. We find that bitmaps are amenable to managing efficient access to large amounts of PMU data. Furthermore, the bitmap-management process will provide decreased access time for data retrieval as well as decreased memory usage. From our experiments, our system is able to achieve approximately 30 times speedup on queries resulting in tuples that are in the database. Conversely, reducing the time taken to answer queries that result in tuples not contained in the database to nearly instantaneous.

This paper presents a systematic approach to detect the frequency of Forced Oscillation (FO) in real-time signals obtained from phasor measurement units (PMUs). It is shown that, preprocessing of the signal by Empirical Mode Decomposition (EMD) technique leads to effective detection. The designed methodology can detect the frequency of FOs even in the presence of multiple mode mixing, whereas the self-coherence method fails to detect. Spectral analysis is performed on the Intrinsic Mode Functions (IMFs) to determine the frequency of FO. To further confirm the sustainability of the oscillation, spectral analysis of time-delayed signals is also performed and the results are discussed. The study is investigated on both simulated signal and real-time PMU signal of North American SynchroPhasor Initiative (NASPI). Two forced oscillations; FO 1 (14.67 Hz) and FO 2 (4.62 Hz) are detected in the NASPI PMU's signal by the proposed approach.