AORCEA - An adaptive operator rate controlled evolutionary algorithm

被引:7
|
作者
Giger, M. [1 ]
Keller, D. [1 ]
Ermanni, P. [1 ]
机构
[1] ETH, Ctr Struct Technol, CH-8092 Zurich, Switzerland
关键词
evolutionary algorithm; strategy parameter control; adaptive operator rates; structural optimization;
D O I
10.1016/j.compstruc.2006.12.002
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
When applying evolutionary algorithms to optimization problems many different strategy parameters have to be set to define the behavior of the evolutionary algorithm itself. To a certain extent these strategy parameter values determine whether the algorithm is capable of finding a near-optimum solution or not. In particular the choice of the different genetic operators and their relative rates is most often based on experience. Furthermore, the operator rates are defined before starting the optimization runs and remain unchanged until the stopping criterion is reached. Controlling the parameter values during the run has the potential of adjusting the algorithm to the problem while solving the problem. This paper investigates an adaptive strategy controlling the rates of arbitrary chosen genetic operators. The control mechanism is based on the state of the optimization by evaluating a success and a diversity measure for each operator. More efficient operators are favored in order to find better solutions with less evaluations. The algorithm is tested with constrained and unconstrained numerical examples and a concrete structural optimization problem is treated. (C) 2007 Elsevier Ltd. All rights reserved.
引用
收藏
页码:1547 / 1561
页数:15
相关论文
共 50 条
  • [1] A novel adaptive hybrid crossover operator for multiobjective evolutionary algorithm
    Zhu, Qingling
    Lin, Qiuzhen
    Du, Zhihua
    Liang, Zhengping
    Wang, Wenjun
    Zhu, Zexuan
    Chen, Jianyong
    Huang, Peizhi
    Ming, Zhong
    INFORMATION SCIENCES, 2016, 345 : 177 - 198
  • [2] Local convergence rate of evolutionary algorithm with combined mutation operator
    Kim, NG
    Won, JM
    Lee, JS
    Kim, SW
    CEC'02: PROCEEDINGS OF THE 2002 CONGRESS ON EVOLUTIONARY COMPUTATION, VOLS 1 AND 2, 2002, : 261 - 266
  • [3] Adaptive Operator Selection With Bandits for a Multiobjective Evolutionary Algorithm Based on Decomposition
    Li, Ke
    Fialho, Alvaro
    Kwong, Sam
    Zhang, Qingfu
    IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2014, 18 (01) : 114 - 130
  • [4] Adaptive composite operator selection and parameter control for multiobjective evolutionary algorithm
    Lin, Qiuzhen
    Liu, Zhiwang
    Yan, Qiao
    Du, Zhihua
    Coello, Carlos A. Coello
    Liang, Zhengping
    Wang, Wenjun
    Chen, Jianyong
    INFORMATION SCIENCES, 2016, 339 : 332 - 352
  • [5] Transgenic: An evolutionary algorithm operator
    do Amaral, Laurence Rodrigues
    Hruschka, Estevam Rafael, Jr.
    NEUROCOMPUTING, 2014, 127 : 104 - 113
  • [6] Many-Objective Evolutionary Algorithm Based on SOM Clustering and Adaptive Operator Selection
    Zhong P.-L.
    Li M.
    He C.
    Chen H.
    Tien Tzu Hsueh Pao/Acta Electronica Sinica, 2022, 50 (08): : 1959 - 1974
  • [7] A classification tree and decomposition based multi-objective evolutionary algorithm with adaptive operator selection
    Huantong Geng
    Ke Xu
    Yanqi Zhang
    Zhengli Zhou
    Complex & Intelligent Systems, 2023, 9 : 579 - 596
  • [8] A classification tree and decomposition based multi-objective evolutionary algorithm with adaptive operator selection
    Geng, Huantong
    Xu, Ke
    Zhang, Yanqi
    Zhou, Zhengli
    COMPLEX & INTELLIGENT SYSTEMS, 2023, 9 (01) : 579 - 596
  • [9] A Clonal Selection Adaptive Local Search Operator for multi-objective optimization evolutionary algorithm
    Li, Yong
    Wang, Yu
    Zhang, Yuxian
    An, Yuejun
    PROCEEDINGS OF THE 2012 24TH CHINESE CONTROL AND DECISION CONFERENCE (CCDC), 2012, : 755 - 757
  • [10] Training and testing a self-adaptive multi-operator evolutionary algorithm for constrained optimization
    Elsayed, Saber M.
    Sarker, Ruhul A.
    Essam, Daryl L.
    APPLIED SOFT COMPUTING, 2015, 26 : 515 - 522