Abstract: The development of electricity networks towards the future smart grids is naturally accompanied by increasing complexity of technical, economical and environmental problems. The new challenges require the development of new techniques and optimization methods, including specific approaches to multiobjective optimization problems.
The presentation will focus on basic and multiobjective optimization methods with application to power system optimization problems. These methods are based on modern Evolutionary Computation (EC) metaheuristics, namely search algorithms based on ordered movement of particles (OMP), such as the Gravitational Search Algorithm (GSA) or the Charged System Search Algorithm (CSSA). As the lecture will emphasize, the performance of these algorithms can be substantially improved using an auxiliary mechanism called “dynamic supervision”, specially designed for escaping from local minima.
At the same time, the increasing number of actors in the electricity industry, along with a more distributed nature of electricity generation and an increasingly active role of consumers in the resource management activities, make the decision making process more and more based on multiobjective approaches. One of the most interesting issues in this context is the optimization problem with conflicting objectives. In this case, beyond the traditional methods such as the weighted sum of objective functions or the bounded objective function, one of the most commonly used technique is based on the principle of Pareto optimality, and can be approached using methods such as Strength Pareto Evolutionary Algorithm (SPEA) or Nondominated Sorting Genetic Algorithm (NSGA). Starting from the principle of Pareto optimality, an original method for handling multiobjective optimization problems using OMP-type search algorithms will be described. The method uses non-dominant sorting and particles clustering based on the rank of the Pareto front where each particle was located.
The implementation of the proposed search algorithms is demonstrated for the case of a classical power systems problem, namely the optimal reactive power compensation or VAR optimization using capacitor banks.

Brief Biography of the Speaker:
Mihai Gavrilas was born in Iasi, Romania on February, 6, 1959. He received the M.Sc. degree from the “Gheorghe Asachi” Technical University of Iasi, Romania, in 1984. Between 1984 and 1988 he worked as a field engineer and then as a design engineer in the field of power station and substation building and design. Since 1988 he has been devoted to education and research at the “Gheorghe Asachi” Technical University of Iasi, where, at present, he is a professor with the Power System department, at the faculty of Electrical Engineering. In 1994 he received the Ph D degree in power systems from the Technical University of Iasi. He is reading courses on Power systems steady state and stability analysis, Intelligent systems application in power systems and Electricity markets. He has a remarkable scientific and educational experience being the author or joint author of 11 books, and over 140 papers in the area of power systems and intelligent systems applications, published in international journals and conference proceedings. He also has a valuable project management experience (project manager or member in the research team) in over 40 research grants sponsored by research organizations and / or research programs with partners from industry. He is an IEEE member (Power and Energy Society, Computational Intelligence Society, Systems, Man and Cybernetics Society) since 1994 and a CIGRE member since 2008. His main research interests are directed towards multi-objective optimization for power systems steady-state operating conditions, power systems dynamics and control, state estimation and observability analysis in power systems, and computational intelligence application in power systems. He participated as a Plenary Speaker in the 9-th WSEAS / IASME International Conference on Electric Power Systems, High Voltages, Electric Machines (POWER 2009), Genova, Italy and the 12-th WSEAS International Conference on Mathematical Methods and Computational Techniques in Electrical Engineering (MMACTEE 2010), Timisoara, Romania.