The discipline studies the main machine learning techniques and their application in the power industry. The following sections and machine learning techniques are considered: supervised learning in classification and regression problems (linear models, support vector machines, decision trees, Bayesian methods, k-nearest neighbors, neural networks, ensemble algorithms), unsupervised learning (clustering algorithms), reinforcement learning and optimization methods (genetic and swarm algorithms), fuzzy logic. Special focus is given to preliminary analysis and data processing approaches.

The discipline studies the basic principles of assuring data security and providing protection against cyber threats for power system facilities. The discipline covers legal and organizational issues, software specifics, technical and algorithmic methods of data protection used to assess the threats and to model the corresponding protection measures. Special consideration is given to the existing legal framework in the sphere of data protection and the relevant industrial standards. The discipline forms deep understanding of the types of data to be protected, the classification of cyber threats at power system facilities, various protection techniques, the principles of their operation and selection criteria. The project is implemented by the winner of the Master's Program Faculty Grant Competition of the Vladimir Potanin Fellowship Program.

The discipline studies the principles of designing urban power grids and provides the detailed consideration of different types of energy consumers, their technological operation modes, power quality requirements. Special consideraion is given to the energy saving technologies and renewable energy sources application in modern power systems.

The discipline highlights the design and operation issues of small-scaled power generation facilities (distributed generation), including hydrocarbon and renewable energy-based ones. Technical and economic issues related to distributed generation are investigated, including distributed generation operation modes, grid connection and parallel operation with bulk power system.

The discipline covers the fundamentals of the technical diagnosis theory with particular focus on high-voltage power equipment. The primary barriers associated with technical diagnosis effective implementation are highlighted with detailed consideration of the existing technologies and tools as well as the relevant methodologies for creating diagnosis and monitoring systems. As part of the discipline, much attention is paid to the basics of designing diagnosis systems, applying nondestructive testing methods and selecting diagnostic parameters for the object of study. As a result of mastering the discipline the students will gain deep understading of the relation between the physical aspects of power equipment operation, aging, external factors and diagnostic parameters.

The discipline is devoted to the application of optimization techniques to solve the critical problems of the power industry. Due to the task-orented nature of the discipline, special focus is given to the optimization methods application, skill-building in setting and solving the optimization problems. Much attention is paid to the information support of the power system control tasks. The application of Matlab, MathCAD and PowerFactory to practical cases of power system optimization and load forecasting is carefully studied. Within the discipline, the principles and methods of taking into account various constraints and external impacts are studied.