Uniform and Orthogonal Evolution Algorithm with Powell Local Search

摘要：

In this paper,a uniform and orthogonal evolution algorithm with Powell local search strategy (UOEA/PL) is proposed based on the Powell direct optimization method and the technique of the uniform and orthogonal experiment design.Uniform design method ensures that original solutions are distributed in the searching space uniformly.Less experiments is needed by using the orthogonal cross-over operation,which accelerates the convergence of the algorithm.Powell direct optimization method greatly improves the local search ability and solution quality.The pre-maturity and local optima can be effectively avoided by the random mutation operator.UOEA/PL algorithm is tested by using the modified benchmark functions.Simulation results shows that UOEA/PL is generally effective and for global optimization problems.

关键词： Powell evolutionary algorithm uniform design orthogonal design

作者: Riming Wang Jiuchao Feng

作者单位: School of Electronic and Information Engineering South China University of Technology Guangzhou,Chin School of Electronic and Information Engineering South China University of Technology Guangzhou,Chin

会议类型: 国际会议

会议名称: 2011 3rd International Conference on Computer and Network Technology(ICCNT 2011)(2011第三届IEEE计算机与网络技术国际会议)

会议地点: 太原

会议语种:英文

页码: 90-94

在线出版日期: 2011-02-26（万方平台首次上网日期，不代表论文的发表时间）