Data Fusion and Data Mining for Power System Monitoring

Data Fusion and Data Mining for Power System Monitoring provides a comprehensive treatment of advanced data fusion and data mining techniques for power system monitoring with focus on use of synchronized phasor networks. Relevant statistical data mining techniques are given, and efficient methods to cluster and visualize data collected from multiple sensors are discussed. Both linear and nonlinear data-driven mining and fusion techniques are reviewed, with emphasis on the analysis and visualization of massive distributed data sets. Challenges involved in realistic monitoring, visualization, and analysis of observation data from actual events are also emphasized, supported by examples of relevant applications. Features Focuses on systematic illustration of data mining and fusion in power systems Covers issues of standards used in the power industry for data mining and data analytics Applications to a wide range of power networks are provided including distribution and transmission networks Provides holistic approach to the problem of data mining and data fusion using cutting-edge methodologies and technologies Includes applications to massive spatiotemporal data from simulations and actual events

Chapter 1Introduction1.1.- Introduction to power system monitoring1.2.- Wide-area power system monitoring1.3.- Data fusion and data mining for power health monitoring1.4.- Dimensionality reduction1.5.- Distribution system monitoring1.6.- Power system data1.7.- Sensor placement for system monitoring1.8.- ReferencesChapter 2Data mining and data fusion architectures2.1.- Introduction2.2.- Trends in data fusion and data monitoring2.3.- Data mining and data fusion for enhanced monitoring2.4.- Data fusion architectures for power system monitoring2.5.- Open issues in data fusion2.6.- ReferencesChapter 3Data parameterization, clustering and denoising3.1.- Introduction: Backgroung and driving forces3.2.- Spatio-temporal data sets projections and spatial maps3.3.- Power system data normalization and scaling3.4.- Nonlinear dimensionality reduction3.5.- Clustering schemes3.6.- Detrending and denoising of power system oscillations3.7.- References Chapter 4Spatio-temporal data mining4.1.- Introduction4.2.- Data mining and knowledge discovery4.3.- Spatio-temporal modeling of dynamic processes 4.4.- Space-time prediction and forecasting4.5.- Space-temporal data mining and pattern evaluation4.6.- ReferencesChapter 5Multisensor data fusion5.1.- Introduction and motivation5.2.- Spatio-temporal data fusion5.3.- Data fusion principles5.4.- Multisensor data fusion framework5.5.- Multimodal data fusion techniques5.6.- Case study5.7.- ReferencesChapter 6Dimensionality reduction and feature extraction and classification6.1.- Background and driving forces6.2.- Fundamentals of dimensionality reduction6.3.- Data-driven feature extraction procedures6.4.- Dimensionality reduction methods6.5.- Dimensionality reduction for classification and cluster validation6.6.- Markov dynamic spatio temporal models6.7.- Sensor selection and placement6.8.- Open problems in nonlinear dimensionality reduction6.9.- ReferencesChapter 7Forecasting decision support systems7.1.- Introduction7.2.- Backgroud: Early warning and decision support systems7.3.- Data-driven prognostics7.4.- Space-time forecasting and prediction7.5.- Kalman flitering approach to system forecasting7.6.- Dynamic harmonic regression7.7.- Damage detection7.8.- Power systems time series forecasting7.9.- Anomaly detection in time series7.10.- ReferencesChapter 8Data fusion and data mining analysis and visualization8.1.- Introduction8.2.- Advanced visualization techniques8.3.- Multivariable modeling and visualization8.4.- Cluster-based visualization of multidimensional data8.5.- Spatial and network displays8.6.- ReferencesChapter 9Emerging topics in data mining and data fusion9.1.- Introduction9.2.- Dynamic spatio-temporal modelling9.3.- Challenges for the analysis of high-dimensional data9.4.- Distributed data mining9.5.- Dimensionality reduction9.6.- Bio-inspired data mining and data fusion9.7.- Other emerging issues9.8.- Application to power system data9.9.- ReferencesChapter 10Experience with the application of data fusion and data mining for power system health monitoring10.1.- Introduction10.2.- Background10.3.- Sensor placement10.4.- Cluster-based visualization of transient performance10.5.- Multimodal fusion of observational data10.6.- References

Arturo Messina earned his PhD from Imperial College, London, UK, in 1991. Since 1997, he has been a professor in the Center for Research and Advanced Studies, Guadalajara, Mexico. He is on the editorial and advisory boards of Electric Power Systems Research, and Electric Power Components and Systems. From 2011 to 2018 he was Editor of the IEEE Trans. on Power Systems and Chair of the Power System Stability Control Subcommittee of the Power Systems Dynamic Committee of IEEE (2015-2018). A Fellow of the IEEE, he is the editor of Inter-Area Oscillations in Power Systems – A Nonlinear and Non-stationary Perspective (Springer, 2009) and the author of Robust Stability and Performance Analysis of Large-Scale Power Systems with Parametric Uncertainty: A Structured Singular Value Approach (Nova Science Publishers, 2009), and Wide-Area Monitoring of Interconnected Power Systems (IET, 2015).