Title :
Research on the optimal combination of ACO parameters based on PSO
Author :
Xie, Xin ; Wu, Peng
Author_Institution :
Sch. of Inf. Eng., East China JiaoTong Univ., Nanchang, China
Abstract :
As ACO´s different value of parameters has impact on its performance. Try to find the optimal combination of parameters can make algorithm have best performance. On the basis of basic model of ACO, we analyzed parameters´ effects on algorithm performance, then proposed “TwoStep” strategy for selecting algorithm parameters: First, determine a reasonable range of each parameter, then introduce fitness function and use PSO to receive optimal combination of parameters. Experiments demonstrate that the proposed strategy can obtain good searching results, which may contribute to the promotion and application of ACO.
Keywords :
particle swarm optimisation; ACO parameters; ant colony optimization; fitness function; optimal combination; particle swarm optimization; two-step strategy; Algorithm design and analysis; Ant colony optimization; Chemicals; Data mining; Dynamic programming; Heuristic algorithms; Job shop scheduling; Performance analysis; Probability; Proportional control; ACO; PSO; optimization; two-step;
Conference_Titel :
Networking and Digital Society (ICNDS), 2010 2nd International Conference on
Conference_Location :
Wenzhou
Print_ISBN :
978-1-4244-5162-3
DOI :
10.1109/ICNDS.2010.5479311