Particle Swarm Optimization and hybrid algorithm applied to generation and demand dispatch

This paper addresses the generation and demand dispatch problem using the Particle Swarm Optimization (PSO) technique and a hybrid algorithm, based on the PSO and the differential evolution algorithm. The problem is formulated in the context of a small grid, whose manager dispatches generation and flexible demand along a time horizon in order to minimize generation costs. Unit commitment of generation units is represented. Power flow equality constraints and inequality constraints due to operational limits are modeled. Two applications, one using the IEEE 30-bus test system, are conducted in order to assess and compare both evolutionary algorithms performance. The importance of comparing stochastic algorithms performance on a statistical base is stated.