Compensación de potencia reactiva en sistemas desbalanceados utilizando algoritmos genéticos

the energy efficiency of an electric system depends in great extent of the reactive power compensation that is carried out. one of the ways most used for reactive is power of capacitor banks. a good selection of these, presupposes three aspects: the selection of the banks capacity, the compensator type to be used and the allocation in the power system. these aspects are treated through independent methods. however, each one depends on the rest and all have an economic implication, which leads to the necessity of their consideration in the optimization problem itself. there are references of methods that consider the three aspects in a same problem, but these present restrictions in their applications for secondary power systems. in the paper, we have developed a method with the use of genetic algorithms, which facilitates a more exact problem formulation. the method has been programmed in software matlab, version 7.8 (r2009a), using programming structures of genetic algorithm and direct search toolbox. the effectiveness is shown in two zstudy cases of systems hotels, belonging to the commercial building consumer sector.