Abstract:

We propose a kinetic-molecular theory optimization algorithm based on crystallization process (C-KMTOA) to solve the problem that the KMTOA is easily stuck into local optimal and low accuracy. We also design a separation operator by simulating the crystallization process, which divides the population into three subgroups: the best individuals, the excellent individuals and the worst individuals. In addition, with the help of guiding operation, the worst individuals can move toward the excellent individuals and the excellent individuals move toward the best individuals, so that the search range is narrowed down to the  optimal solution quickly. Experimental results show that the proposed algorithm is superior to the GA, DE, QPSO, and KMTOA algorithms in terms of optimization precision and dynamic performance.

Key words: