Optimal Coordination and Penetration of Distributed Generation with Multi Shunt FACTS Compensators Using GA/Fuzzy Rules

Abstract

In an open electricity market, every consumer will be able to buy his own electricity from any source desired with the result that the unplanned power exchanges are increasing. In order to cope with these kind of problems and increase usable power distribution capacity, distribution generation technology (DG) and Flexible AC transmission systems (FACTS) where developed and introduced to the market. Optimal placement and sizing of distribution generation is a well-researched subject which in recent years interests many expert engineers. Efficient placement and sizing of distribution generation (DG) in practical networks can result in minimizing operational costs, environmental protection, improved voltage regulation, power factor correction, and power loss reduction (Mendez et al., 2006). DG is defined as any source of electrical energy of limited size interconnected to the distribution system. DG technologies include photovoltaic systems, wind turbines, fuel cells, small micro-sized turbines, sterling-engine based generators and internal combustion engine-generators (Vovos et al., 2007). In practical installation and integration of DG in power system with consideration of FACTS devices, there are five common requirements as follows (Mahdad et al., 2007): What Kinds of DG and FACTS devices should be installed? Where in the system it should be placed? How to estimate economically the number, optimal size of DG and FACTS to be installed in a practical network? How to coordinate dynamically the interaction between multiple DG, FACTS devices and the network to better exploit the DG and FACTS devices to improve the index power quality? How to review and adjust the system protection devices to assure service continuity and keep the index power quality at the margin security limits? The global optimization techniques known as genetic algorithms (GA), simulated annealing (SA), tabu search (TS), and evolutionary programming (EP), which are the forms of probabilistic heuristic algorithm, have been successfully used to overcome the non-convexity 15