Abstract:

Aiming at the complex multi-objective problems, a multi-objective particle swarm optimization algorithm based on multi-population dynamic cooperation was proposed. In this algorithm, multiple populations were set up to search independently at the same time, thereby effectively improving the search ability of the algorithm. In addition, in order to further ensure the diversity of the population, the algorithm divided the population into two sub-populations by dynamic clustering strategy, and changed the updating mode of subpopulations. Furthermore, dynamic learning samples and differential mutation were used to further prevent the algorithm from falling into local optimum. A series of standard test functions were simulated to verify the effectiveness of the algorithm on multi-objective problems. In addition, comparing the algorithm with five existing algorithms, the results showed that the algorithm has obvious advantages in diversity and convergence.

Key words: multi-objective optimization, particle swarm optimization, multi-population, dynamic clustering, dynamic learning samples, differential mutation