Abstract: This paper proposes a hybrid algorithm (Powell-NLCPSO) based on the Powell search method and the local constriction approach particle swarm optimization with nonlinear varying inertia weight for unconstrained optimization. PowellNLCPSO is very easy to implement in practice since it does not require gradient computation. The modification of both the Powell search method and the particle swarm optimization intends to produce faster and more accurate convergence. The main purpose of the paper is to demonstrate how the standard particle swarm optimizers can be improved by incorporating a hybrid strategy. The computational results of a suit of 20 test function problems taken from the literature show that the hybrid PowellNLCPSO approach outperforms other three relevant search techniques (i.e., the original PSO, the guaranteed convergenceparticle swarm optimization (GCPSO) and hybrid NMPSO) in terms of solution quality and convergence rate. In a later part of the comparative experiment, the PowellNLCPSO algorithm is compared to various most uptodate cooperative PSO (CPSO) procedures appeared in the literature. The comparison report still largely favors the PowellNLCPSO algorithm in the performance of accuracy, robustness and function evaluation. As evidenced by the overall assessment based on computational experience, the new algorithm has been demonstrated to be extremely effective and efficient at locating bestpractice optimal solutions for unconstrained optimization.

Key words: Powell search method, particle swarm optimization, unconstrained optimization