New evolution algorithm based on the standard particle swarm optimization

The particle swarm optimization (PSO) is an alternative for global optimization. A standard for PSO (SPSO) was defined which took into the latest developments, and was used as a baseline for performance testing of improvements. In SPSO, however, particles need to search the optimal solution in the constraint space, and the item velocity in PSO makes particles difficult to adjust themselves to meet those complicated constraints. A new PSO without the item velocity based on SPSO is proposed in this article. The new algorithm inherits the capability of SPSO with fast convergence and high accuracy. This research proves that the convergence process of PSO has nothing to do with the velocity and the proposed method modified simple PSO (msPSO) can converge. The experiments show that msPSO is able to achieve a good result.