Orthogonal methods based ant colony search for solving\ud continuous optimization problems

Article English OPEN
Hu, X. ; Zhang, J. ; Li, Y. (2008)
  • Publisher: Springer
  • Related identifiers: doi: 10.1007/s11390-008-9111-5
  • Subject: QA75 | QA76
    acm: MathematicsofComputing_NUMERICALANALYSIS | ComputingMethodologies_ARTIFICIALINTELLIGENCE

Research into ant colony algorithms for solving continuous optimization problems forms one of the most\ud significant and promising areas in swarm computation. Although traditional ant algorithms are designed for combinatorial\ud optimization, they have shown great potential in solving a wide range of optimization problems, including continuous\ud optimization. Aimed at solving continuous problems effectively, this paper develops a novel ant algorithm termed "continuous orthogonal ant colony" (COAC), whose pheromone deposit mechanisms would enable ants to search for\ud solutions collaboratively and effectively. By using the orthogonal design method, ants in the feasible domain can explore\ud their chosen regions rapidly and e±ciently. By implementing an "adaptive regional radius" method, the proposed\ud algorithm can reduce the probability of being trapped in local optima and therefore enhance the global search capability and accuracy. An elitist strategy is also employed to reserve the most valuable points. The performance of the COAC is\ud compared with two other ant algorithms for continuous optimization of API and CACO by testing seventeen functions\ud in the continuous domain. The results demonstrate that the proposed COAC algorithm outperforms the others.
  • References (64)
    64 references, page 1 of 7

    [1] Deneubourg J L, Aron S, Goss S, Pasteels J M. The selfspeed and avoid trapping in local optima and converge organizing exploratory pattern of the Argentine ant. Journal prematurely. (b) The best region becomes a region of Insect Behavior, 1990, 3: 159{168. with a larger radius in the current iteration. (c) If the [2] Goss S, Aron S, Deneubourg J L, Pasteels J M. Self-

    organized shortcuts in the Argentine ant. Naturwisradius is smaller than the prede¯ned value, it will be senschaften, 1989, 76(12): 579{581. reset to a random value that may be larger than the [3] Dorigo M, StuÄtzle T. Ant Colony Optimization. the MIT previous one. Press, 2003.

    For unimodal functions, the radiuses are gradually [4] Dorigo M, Gambardella L M. Ant colony system: A cooper-

    ative learning approach to the traveling salesman problem. reduced except for f4 in Fig.7. For multimodal func- IEEE. Trans. Evol. Comput., 1997, 1(1): 53{66. tions and f4, whose global optima are harder to ¯nd, [5] Toth P, Vigo D. The Vehicle Routing Problem. SIAM Monothe radiuses °uctuate with step changes. Each step graphs on Discrete Mathematics and Applications, Philadel-

    phia, Society for Industrial & Applied Mathematics, 2001. jump accompanies a great reduction of the function [6] Gambardella L M, Taillard E¶ D, Agazzi G. MACS-VRPTW: values, where large radiuses can help jump out of lo- A Multiple Ant Colony System for Vehicle Routing Problems cal optima. with Time Windows. New Ideas in Optimization, Corne

    3) As the graphs are drawn by radiuses and function D, Dorigo M, Glover F (eds.), London, McGraw Hill, 1999,

    pp.63{76. values of the best results, the radiuses are unchanged [7] Zhang J, Hu X M, Tan X, Zhong J H, Huang Q. Impleif no better results have been found. Hence, horizon mentation of an ant colony optimization technique for job lines are shown in the graph, such as the radiuses of shop scheduling problem. Transactions of the Institute of f1 and f3 etc. Measurement and Control, 2006, 28(1): 1{16.

    [8] Zecchin A C, Simpson A R, Maier H R, Nixon J B. Paramet-

    It can be seen from the graphs that the sizes of the ric study for an ant algorithm applied to water distribution radiuses control the accuracy of the resulting function system optimization. IEEE Trans. Evol. Comput., 2005, 9: value. The smaller the radiuses, the higher the accu- 175{191. racy of the result is achieved. [9] Parpinelli R S, Lopes H S, Freitas A A. Data mining with

    Comput., 2002, 4: 321{332. 7 Conclusions [10] Sim K M, Sun W H. Ant colony optimization for routing and

  • Similar Research Results (2)
  • Metrics
    No metrics available
Share - Bookmark