Intelligent Data Analytics for Power Apparatus Health Monitoring

Intelligent Data Analytics for Power Apparatus Health Monitoring: AI and Machine Learning Paradigms reviews key implementations of machine learning and data analytics techniques for the optimization of digital power transformers. The work addresses health monitoring fully across the constitutive structure of modern transformers, with coverage of DGA-based intelligent data analytics, transformer winding, bushing and arrestor health monitoring, core, conservator, and tank and cooling systems. Chapters address advanced AI / ML methods including deep convolutional neural network, fuzzy reinforcement learning, modified fuzzy Q learning, gene expression programming, extreme-learning machine, semi-supervised & unsupervised ELM, proximal support vector machine, deep learning neural network, long short-term memory, and network and deep learning neural networks. Primarily intended for researchers and practitioners, the book speeds and simplifies the diagnosis and resolution of health and condition monitoring queries using advanced techniques, particularly with the goal of improved performance, reduced cost, optimized customer behavior and satisfaction, and ultimately increased profitability. Discusses pathways to apply machine learning techniques and advanced artificial intelligence algorithms in transformer diagnostics Provides demonstrator codes in R, WEKA and MATLAB supporting direct implementation of intelligent data analytics for each power transformer component Demonstrates realistic methods for the step-by-step implementation of advanced AI / ML health monitoring, including convolutional neural network, deep learning, and extremely learning machine approaches