Metaheuristicas aplicadas ao planejamento da expansão da transmissão de energia eletrica em ambientes de processamento distribuido




In this work, several combinatorial metaheuristics are developed for solving the transmission expansion planning problem of electric power systems that is analysed considering the static and long-term approach, e.g. a parallel version of the simulated annealing methodology and several parallel versions of a genetic algorithms, besides an asynchronous team which agents are variants of these metaheuristics. All of these versions are initialized by an asynchronous team of constructive heuristic algorithms, executed in a distributed processing environment, composed of a heterogeneous network of SUN workstations, SunOS, with PVM parallel processing library. Several tests are effectuated for the systems: Garver (6 busses/15 branches); brazilian South (46 busses/79 branches), brazilian North/Northeast (87 busses/179 branches) and the colombian system (93 busses/155 branches). The results show the efficiency of the proposed methodologies when compared to the serial versions of each metaheuristic isolatedely, as well as a significative reduction on the processing times.


otimização combinatoria sistemas de energia eletrica energia eletrica processamento paralelo (computadores)

Documentos Relacionados