An effective and efficient two stage algorithm for global optimization

Abstract

A two stage algorithm, consisting of gradient technique and panicle swarm optimization (PSO) method for global optimization is proposed. The gradient method is used to find a local minimum of objective function efficiently, and PSO with potential parallel search is employed to help the minimization sequence to escape from the previously converged local minima to a better point which is then given to the gradient method as a starting point to start a new local search. The above search procedure is applied repeatedly until a global minimum of the objective function is found. In addition, a repulsion technique and partially initializing population method are also incorporated in the new algorithm to increase its global search ability. Global convergence is proven, and tests on benchmark problems show that the proposed method is more effective and reliable than the existing optimization methods. © Springer-Verlag Berlin Heidelberg 2006.