2009•Shuxue de shijian yu renshiRequires access
Ant Colony Optimization Genetic Hybrid Algorithm
Xiaoru Zhang
Open publisher page 4 citations
Abstract
Ant colony optimization genetic hybrid algorithms are proposed for obtaining optimal value of discrete space and continuous space.The framework of hybrid algorithm solving the traveling salesman problems is genetic algorithm,and based on the properties of pheromone in ant colony algorithm the crossover operation is given.Four mutation strategies are put forward using the characteristic of traveling salesman problems.The hybrid algorithm with 2-opt local search can effectively find better minimum beyond premature convergence.Compare with the simulated annealing algorithm,the standard genetic algorithm and the standard ant colony algorithm,all the 4 hybrid algorithms are proved effective.Especially the hybrid algorithm with strategy D is a simple and effective better algorithm than others.The framework of hybrid algorithm solving the continuous space optimum problems is ant colony algorithm,and the crossover operator step and mutation operator step are added.The efficiency of the hybrid algorithm is verified by test function.