Short description:
Energy Systems is an interdisciplinary educational track at Skoltech with focus on Power Systems Control and Optimization, Smart Grids, Energy Conversion, Power Electronics and Energy Markets. The broad educational program builds upon core and target area courses taught by faculty members of Center for Energy Science and Technology and elective courses delivered by members of other CREIs including Center for Computational and Data-Intensive Science and Engineering (CDISE) and Skoltech’s Space Center (SSC). Following major trends in energy sector the program also delivers an interdisciplinary vision of energy systems as coupled electric/heat/gas infrastructures that is unique for the universities in Russia and beyond. Other core aspects of the program are industry immersion and the entrepreneurship and innovation component. Skoltech offers international environment and the opportunities for students to visit foreign universities and international conferences.
The education program consists of core and elective courses with recommendations for two tracks – Energy systems and Energy transition and ESG as detailed below as detailed below:
Core courses
Course instructor: Prof. Elena Gryazina, Sergey Parsegov, Oleg Khamisov Description: This course covers power systems analysis & operations, including fundamentals (balanced three-phase power) steady-state analysis (power flow), state estimation, operation (optimal power flow), security (contingency analysis and security constrained optimal power flow), power system dynamics (frequency and voltage control) and challenges and trend of future power systems. After successfully completing this course, the student will be capable of analyzing the technical and economic operation of an electric energy system.
Course instructor: Prof. Henni Ouerdane Description: Classical equilibrium thermodynamics is a theory of principles, which provides a framework to study means to produce motive power and useful heat, crucial for our everyday life. It is a pillar of any serious physics and engineering curriculum. This graduate course provides the students from possibly diverse backgrounds with the theoretical concepts that underlie the physics of energy conversion at the heart of heat engines operation, including chemical processes, and the specific knowledge of energy technologies in use nowadays. Covering some of the main real-world technologies for the generation of electric/mechanic, heating and cooling power: boilers, steam and organic Rankine cycles, gas turbines, internal combustion engines, heat pumps and chillers to name a few, students will learn to critically analyze and assess these technologies to improve their performance and imagine innovative and commercially viable solutions to energy problems, accounting for costs and environmental aspects like pollutants formation and their abatement. Essential notions which are taught include: energy conversion; heat transfer; work; first and second principles; working fluids and thermoelastic coefficients; chemical reactions; thermodynamic cycles; motors and refrigerators: engines and heat pumps; sources of irreversibility; finite-time thermodynamics. Time permitting, notions of kinetic theory and statistical thermodynamics may be briefly introduced. The course is organized around the learning of essential concepts and an awareness development of current energy technologies. It is based both on “teaching with lecture” and “teaching with discussions” methods. In addition to home assignments and project, students will solve problems during tutorials and discuss their solutions.
Course instructors: Prof. Elena Gryazina, Prof. Henni Ouerdane, Prof. Vladimir Terzija, Prof. Federico Ibanez This research seminar is the general meeting for faculty, researchers and master and PhD students of Energy Systems programs. The seminar takes place every week during Terms 2(6)-3(7)-4(8). Master students must attend the seminar at least for one academic year but welcome to attend during two years. PhD students are welcome to attend the seminar during all years of studies but can gain no more than 6 credits in total. The seminar consists of faculty lectures, invited lectures of top scientists in their research field as well as students’ reports on their own or examined papers
Course instructors: Prof. David Pozo Description: The course will introduce the students to power system economics. After covering fundamentals of microeconomics, main types of electricity markets and regulation will be discussed including the Russian market. Economic dispatch and Optimal Power Flow with Locational Marginal Pricing will also be covered. The lectures will be supplemented by homeworks utilizing PowerWorld simulation package, a laboratory exercise investigating gaming in power markets and group mini-projects.
Recommended electives in power systems and power electronics
Course instructor: Prof. Federico Ibanez Description: The course provides an overview of the latest achievements in power electronics. The main purpose of the course is to analyze different circuit topologies, to understand how they work and which are their benefits and limitations. The course starts with reviewing the basics in electric circuit theory, and then, it introduces different kind of semiconductor devices such as diodes, thyristors and transistors. After this, power electronics circuits are presented: rectifiers, DC-DC converters and inverters. The course gives the tools to analyze any kind of power converters, and provides different examples related with microgrids and energy storage applications. It has three parts: lectures, home tasks and experimental activities in the lab. By the end of the course, the students should be able to analyze a power converter, to simulate it and to understand the possible applications.
Course instructor: Prof. Federico Ibanez Description: Power systems around the world are undergoing a period of unprecedented change. A typical 20th Century power system was characterized by unidirectional flow of power from a limited number of large controllable power stations to a highly predictable demand. There was no energy storage so that at any time generation had to be equal to demand and the infrastructure utilization rates were low (about 55% for generation, 30% for transmission and even lower for distribution). Generally planning and controlling such a system was relatively straightforward as it was based around principles of deterministic hierarchical control, usually based on (N-1) reliability criterion. On the other hand the emerging 21st Century power system is characterized by bidirectional flows between a very large number of uncontrollable and stochastic generators (usually, but not always, renewable ones such as wind or solar) and stochastic and often poorly-predictable demand. Demand ceases to be predictable as it consists of consumers equipped with smart meters and wind/solar generators hence possibly becoming net generators – so-called prosumers. Increased penetration of energy storage, both stationary and mobile due to a take-up of electric vehicles, offers buffering possibilities in dispatch (generation does not have to be equal to demand at any time). Controlling such a power system is the main research challenge in power systems and it is made possible by latest advances in ICT (Information and Control Technology), communication networks, Internet, GPS, sensors, etc. However it requires new tools and methodologies, the Smartgrid course will give the basis of this new grid scenario.
Course instructor: Prof. Vladimir Terzija
Course instructor: Prof. Dmitry Titov Recommended electives in data science and computational engineering Course instructor: Prof. Elena Gryazina Description: The aim of this is to recap the basic topics that you are expected to get at bachelor level. If you do not feel confident with basic matrix manipulations, integration and differentiation – you should definitely take this course; fluency in these is a must for an educated engineer. At the first day of studies you’ll take the preliminary exam to make final decision regarding taking this course.
Students will learn how to analyze strategies in the field of digitalization of power grid companies and provide an expert assessment of the chosen strategy, taking into account the technological maturity of the company, as well as other internal and external factors.
Course instructors: Prof. Mikhail Belyayev, Prof. Maxim Panov (Professors of CDISE) Description: The course gives an introduction to the main topics of modern data analysis such as classification, regression, clustering, dimensionality reduction, reinforcement and sequence learning, scalable algorithms. Each topic is accompanied by a survey of key machine learning algorithms solving the problem and is illustrated with a set of real-world examples. The primary objective of the course is giving a broad overview of major machine learning techniques. Particular attention is paid to the modern data analysis libraries which allow solving efficiently the problems mentioned above.
Course instructor: Prof. Andrey Somov (Professor of CDISE) Description: In the last decade the Internet of Things (IoT) paradigm has slowly but steadily and increasingly permeated what researchers and engineers study and build. The term “Internet of Things” doesn’t have a single definition and people today often use it to interchangeably refer to Wireless Sensor Network (WSN), Machine-to-Machine (M2M), Web of Things (WoT) and other concepts. The focus of this course is to learn about these technologies that will be extending the Internet as we know it and use it today, to interconnect not only people and computers but also sensors and associated objects. The course will be divided into two strongly coupled parts. The first part of the course covers the IoT ‘pillar’ technologies, i.e. embedded systems, wireless sensor networks, semantic technologies, and theory behind them while the second part will have a special focus on IoT development, i.e. IoT apps, open platforms, sensors and actuators, software/middleware. Apart from covering the theory behind the IoT and “how to connect things to the Internet”, the course will therefore also engage the students to demonstrate the feasibility of simple IoT real applications and will challenge them to improve their applications through the use of cognitive technologies and cloud computing.
Course instructor: Prof. Elena Gryazina Description: The course is devoted to optimization methods and optimization problems design with a special attention to those motivated by data science, engineering and industrial applications. The course starts with a brief reminder of the foundations of convex analysis. Then we discuss zero, first and second order methods with a special focus on their efficient implementation. We distinguish between various problem classes, discuss suitable methods for every class. The problem formulation and its proper reformulation is the critical issue for an optimizer. We’ll learn about optimization models and convex relaxations. Special attention will be addressed to Linear Matrix Inequalities (LMI) that arise in optimization problem formulations. One of the home assignments is devoted to understanding the constraints of performance for different software packages. At the last part of the course we move further to advanced first order optimization methods such as proximal mirror descent and extra gradient methods and discuss how to utilize problem structure (e.g. sparsity and separability) to speed up the methods. Within engineering and practical sections, we show how to use the methods above to crack convex and non-convex problems arises in engineering, energy systems, machine learning and related fields.
Course instructor: Prof.Evgeny Burnaev Description: Recommended electives in natural sciences Course instructor: Prof. Henni Ouerdane Description: Classical thermodynamics is useful to describe equilibrium states, while non-equilibrium states and irreversibility characterize real physical processes. If one is interested in actual processes at work during energy conversion, a classical thermodynamic description of equilibrium states is insufficient as it yields very incomplete information on the processes. Irreversible thermodynamics accounts for the rates of physical processes, and provides relationships between “measurable quantities” such as transport coefficients. This graduate course, which constitutes the natural continuation of the course Energy Systems Physics & Engineering, provides the students with basic knowledge of out-of-equilibrium and finite-time thermodynamics, which describe irreversible processes that routinely take place in physical systems and permits a fine understanding of the processes ensuring energy conversion. Thermoelectric generators serve as the main example to illustrate in a simple fashion the out-of-equilibrium formalism, and other systems such as, e.g., solar cells are studied. Essential notions which are taught include: Onsager’s approach to linear nonequilibrium thermodynamics; coupled transport theory; Boltzmann equation; thermal conductivity; electrical conductivity; electrochemical potential in solid-state systems; force-flux formalism and its application to thermoelectric systems; device optimization modelling accounting for dissipative coupling to heat reservoirs; solar energy conversion. The course is organized around the learning of essential concepts and an awareness development of current energy technologies. It is based both on “teaching with lecture” and “teaching with discussions” methods. In addition to home assignments and project, students will solve problems during tutorials and discuss their solutions. Recommended electives for track “Energy transition and ESG” Course instructor: Prof. Mikhail Belyaev Description:Computer Vision is one of the most rapidly evolving subfields of Data Science with many applications, e. g. in autonomous driving and healthcare, among others. This course is designed to provide a comprehensive systematic introduction to the field. We’ll start with the recognition of some simple object elements such as corners and edges and then proceed to the detection of more complex local features. All major problem statements such as image classification, object detection and segmentation as well as the corresponding classical algorithms will be covered within the course. Finally, we’ll briefly introduce convolutional networks and discuss key deep learning architectures for the same set of problems. We’ll extensively use Python and CV & image analysis libraries scikit-image and OpenCV during hands-ons and homeworks. The final grade will be calculated using the results of three homeworks (20% each) and the final project (40%).
Course instructor: Prof. Dmitry Dylov Description:In the computational era of everything, imaging has not become an exception. Computational algorithms allow both to extract valuable information from a scene and to improve the very sensor that forms the image. Today, computational and image processing enhancements became integrable parts of any digital imager, be it a miniature smartphone camera or a complex space telescope. This crash course is designed as a prerequisite for those students who would like to venture into the field of Computer Vision. We will cover foundational mathematical equations that are involved in the image formation and in the geometric projection principles. The concept of Point Spread Function that distorts the object will be explained on particular examples and will be experimented with for the tasks of image reconstruction and denoising. Image processing will be covered with an emphasis on the Python libraries to be used in the rest of the imaging-related courses on the DS/IST tracks (openCV and others). A basic DSLR photo camera will be considered as a model for understanding Fourier Imaging and Filtering methods in a laboratory exercise. Hands-on tutorials on how to select a camera and a lens for your machine vision application will be provided. The theory of color and stereo light-field cameras will be covered using the models of commonplace Bayern RGB sensors; as well as state-of-art spectral and multi-lens imagers. The course will consist of three theoretical lectures riffled by three graded in-class laboratory coding sessions on the subjects covered in the theoretical lectures. 100% attendance is mandatory. There will be a single in-class exam during the evaluation week and no homework.
Course instructor: Prof. Dmitry Krasnikov Description:The present course aims to provide a concise but comprehensive introduction to a multidisciplinary field of catalysis. As it affects more than 90% of the chemical industry and generates up to 30 % of the GDP, catalysis employs models and methods from a wide variety of disciplines: from physical chemistry to solid-state physics, from quantum chemistry to hydrodynamics. Thus, two strategies are usually used for teaching: separate educational programs (>2 years) or brief discussion during one or two lectures max. Here we wish to propose a third way – a 3 credit course with problem-oriented education. During the course, the students will learn basic principles and concepts of catalysis and develop a team project on a particular and demanding scientific problem. The latter will be defended during the discussion seminar with other student teams acting as opponents and reviewers. Totally 19 hours of lectures, 9 hours of exercises and 3hours of discussion work. During the courses each student is supposed to take part in the team project that will be concluded with a 15 min presentation. Other core courses and activities Master thesis Skoltech professors offering Master projects in Energy Systems are: Assistant Prof. Elena Gryazina Assistant Prof. Federico Martin Ibanez Assistant Prof. Henni Ouerdane Assistant Prof. Vladimir Terzija Additional information can be found here: The program is managed by:
– Have a good understanding of the fundamental issues and challenges of ML: data,
model selection, model complexity among others;
– Have an understanding of the strengths and weaknesses of many popular ML
approaches;
– Appreciate the basic underlying mathematical relationships within and across
ML algorithms and the paradigms of supervised and unsupervised learning.
– Be able to design and implement various machine learning algorithms in a range of
real-world applications.