Small Population-Based Particle Swarm Optimization for Short-Term Hydrothermal Scheduling

Abstract

A small population-based particle swarm optimization (SPPSO) approach is presented to solve the problem of short-term hydrothermal scheduling (STHS). In the proposed approach, a novel mutation operation that selects the flying guides for each individual is employed to enhance the diversity of the small population. A DE algorithm is employed as an acceleration operation to accelerate the convergence of the approach in case that the optimal result has no significant improvements after several iterations. A migration operation is adopted to keep the swarm's crowding diversity above a desired level. In addition, a special repair procedure, instead of the penalty function approach, is applied to handle the complex equality constraints of STHS. The effectiveness of the approach is demonstrated through three hydrothermal test systems published in literature. The results are also compared with those obtained by other evolutionary methods. The fuel cost as well as other performance of the proposed approach has been found to be quite impressive. It is shown that the SPPSO approach can provide a better solution at lesser computational time and effort.