Hybrid adaptive particle swarm optimization based on average velocity

In order to deal with the problems of the low convergence rate and tending to jump into the local optimum in the traditional particle swarm optimization, a hybrid particle swarm optimization is proposed based on the average velocity. A definition of average velocity is presented to characterize the degree of the activity of particle swarm. The inertial weight and acceleration factors are adjusted by this definition. A switching simulated annealing algorithm and the updating equations of annealing temperature are designed, such that all the particles can converge into the global optimum faster and jump out of the local minimum easily. The experiments of searching optimization of three typical functions are given, and the results show the effectiveness of the proposed algorithm.