Units:

12

Description:

Due to power systems industry deregulation, power systems industry needs professionals with multidisciplinary background in energy technologies, electrical engineering, economics and material science among others. While theoretical foundation in these areas is of highest importance, the industry and the research community highly depend on computer simulations and optimization of very large electrical power networks on different time scales both off-line and in real time. Computer simulation, analysis and optimization are usually the first steps in verifying conceptual designs but frequently, in this fast changing industry, existing tools are quickly outdated. Understanding power systems algorithms and being able to program them allows power systems professionals to produce adequate solutions. This course is about understanding current algorithms, their purpose and shortcomings, and how to modify them to fit problem variations. The course is targeted at senior year students with interests in electrical power systems and graduate students doing research in power and energy areas.

The foundation for this course includes linear algebra, basic dynamic systems, usual programming skills and power systems concepts taught as a part of any curriculum with a focus on power systems. After a short review of linear algebra and dynamic systems numerical simulations, fundamental steady state and dynamic power systems problems are discussed from their algorithmic and programming aspects. The steady state problems include power flow (AC, Fast Decoupled, DC), optimal power flow (ACOPF, DCOPF, ED), unit commitment (Dynamic Programming, MINLP, MINLP), demand forecasting and long term planning. Standard steady state algorithms are extended to handle Load Tap Changers, Phase Shifters and Fast AC Transmission Systems. Dynamic problems include synchronous and asynchronous machines modeling and stability and photovoltaic power generation. Extra attention is paid to dynamics of islanded smart grid operations and stability. Since power system networks are very large, sparse matrix and parallel programming techniques are explored.

While students can use a programming language of their choice, Matlab and Simulink are used for teaching and algorithm analysis and benchmarking. The focus is on reusable and easy to modify code programming techniques. Procedural and Object Oriented programming methodologies are strongly encouraged and to some extent enforced.

Last Modified: 2018-11-13 3:19PM

Semesters offered:

• Spring 2019
• Spring 2011
• Spring 2009