APPLICATION OF PSO ALGORITHM COMBINED WITH LÉVY FLIGHTS FOR DISTRIBUTED GENERATION ALLOCATION WITH THE OBJECTIVE OF REDUCING ACTIVE LOSSES
Distributed generation (DG) has several benefits for the electrical system, such as the reduction of losses and the increase in the network's voltage profile, in addition to postponing investments in network reinforcements. DG also brings benefits to the environment, especially if the energy source is considered clean, such as the sun, wind, biomass, hydraulics, among others. The benefits of DG can be maximized when allocation and scaling are optimal. This optimization problem is known as Optimal Allocation and Scaling of GD (ADOGD). Due to the problem multimodal characteristic, search and optimization may fail, as there are many attractive solutions. This dissertation presents a new meta-heuristic that associates PSO (Particle Swarm Optimization) with Lévy's flights to solve the ADOGD problem, identified as LFPSO. The Lévy's flights help explore new regions in the search process, without compromising the PSO's characteristic. The tests carried out on the 34 and 69 bus showed that the LFPSO is superior to the PSO in the convergence process. The results obtained by the LFPSO, for the allocation and sizing of three DGs, point to a reduction of 94.4 and 97.9% of active losses, respectively for the 33 and 69 bus systems.