Improved gray-encoded evolution algorithm based on chaos cluster for parameter optimization of moisture movement

Abstract

To improve computational precision for parameter optimization of the van Genuchten model in simulating moisture movement in environment protection, an improved gray-encoded evolution algorithm based on chaos cluster is proposed, in which an initial population is generated by chaotic mapping, and the searching range is automatically renewed with the excellent individuals by chaos cluster operation. Its efficiency is verified experimentally. The results indicate that the absolute error by the improved gray-encoded evolution algorithm based on chaos cluster decreases by 7.52% and 40.40%, respectively, and the relative error decreases by 12.65% and 49.95%, respectively, compared to those by the standard binary-encoded evolution algorithm, and the particle swarm optimization algorithm. Improved gray-encoded evolution algorithm based on chaos cluster has higher precision and it is good for the global optimization in the practical parameter optimization in environment system.