A GA-based initialization of PSO for optimal APFS allocation in water desalination plant

Recently, many researches have applied the artificial intelligence techniques to solve the active power filters (APFs) allocation and sizing problem. Some researchers used genetic algorithm (GA), while others used particle swarm optimization (PSO) technique regardless of the shortcomings of each technique. Nowadays, many researches are interested in proposing hybrid techniques to overcome both techniques shortcomings. This paper proposes a new technique that depends basically on forcing PSO to start from initial solutions that guarantee feasible domain obtained using GA. Thus, PSO will be able to define the global optimal solution avoiding the long processing time associated with GA. The proposed technique is investigated to solve the allocation and sizing (AAS) of APFs for a sea water desalination plant as a practical case study. The results are demonstrated and some concluding remarks are provided.