Семинар за рачунарство и примењену математику, 9. фебруар 2016.

• 08. Фебруар, 2016
• Коментари (0)

Наредни састанак Семинара биће одржан у уторак, 9. фебруара 2016. у сали 301ф Математичког института САНУ са почетком у 14:15 часова.

Предавач: Рака Јовановић, Институт за физику, Земун и Qatar Environment and Energy Research Institute (QEERI) Hamad bin Khalifa University

Наслов предавања: DIFFERENT APPROACHES FOR SOLVING THE PROBLEM OF MAXIMAL PARTITIONING OF GRAPHS WITH SUPPLY AND DEMAND

Апстракт: In recent years there has been a growing interest for the problem of the maximal partitioning of graphs with supply and demand (MPGSD), due to its close connection to electrical distribution systems, especially in the context of smartgrids. To be more precise, the MPGSD can be effectively used for the optimization of self-adequacy of interconnected microgrids. The focus of this talk is on several approaches for solving the MPGSD, and its variations. The first one focuses on finding good quality approximate solutions, in reasonable time, for very large graphs through a multiheuristic approach. Secondly, the ant colony optimization (ACO) metaheuristic is applied to MPGSD, and its version with limited capacity. This approach manages to find near optimal solutions for medium size graphs. With the goal of being able to find confirmed optimal solutions for arbitrary graphs a mixed integer programming (MIP) model is defined. The developed MIP formulation can generally be applied only for small problem instances, but proves to be very suitable for highly sparse graphs where it even manages to outperform ACO.