The Institute for Energy Systems, Energy Efficiency and Energy Economics at TU Dortmund offers a block course on distributed optimization in June/July 2021. 

Optimization algorithms play a crucial role for the operation, planning and design of energy systems. Moreover, in view of the ever increasing size of models and the increasing importance of multi-agent solutions for transmission and distribution systems, the course provides an introduction in the state of the art of distributed algorithms for solving unconstrained and constrained optimization problems. Such problems do not only arise in energy systems but also in sensor fusion and control problems. In order to foster the learning outcome, the method-oriented parts of the course will be combined with programming sessions in Julia.

A limited number of slots for participation of external students/doctoral students is available.

Suggested Pre-knowledge:

• Basic programming skills
• Basic knowledge of nonlinear programming (a brief recap will be given in the course)

Content:

• State-of-the-art centralized optimization algorithms such as the gradient method, interior point and sequential quadratic programming methods - Block 1
• Distributed optimization algorithms such as dual decomposition, the alternating direction method of multipliers (ADMM) and decomposition techniques for nonlinear programming - Block 2

Participants will implement all algorithms considered in the emerging scientific programming language Julia

Further information: DOPT: Practical Distributed Optimization in Julia - IE3 - TU Dortmund (tu-dortmund.de)