计算机工程 ›› 2012, Vol. 38 ›› Issue (5): 202-204,207.doi: 10.3969/j.issn.1000-3428.2012.05.062

• 人工智能及识别技术 • 上一篇    下一篇

自适应差分进化算法及其在参数估计中的应用

王海伦1,余世明2,郑秀莲1   

  1. (1. 衢州学院电气与信息工程学院,浙江 衢州 324000;2. 浙江工业大学智能信息系统研究所,杭州 310014)
  • 收稿日期:2011-08-25 出版日期:2012-03-05 发布日期:2012-03-05
  • 作者简介:王海伦(1979-),女,讲师、硕士,主研方向:智能优化,无线传感器网络;余世明,教授、博士;郑秀莲,副教授、硕士
  • 基金项目:
    2010年度科技部国家重点新产品计划基金资助项目(2010GRC20113);衢州学院中青年骨干教师基金资助项目

Adaptive Differential Evolution Algorithm and Its Application in Parameter Estimation

WANG Hai-lun 1, YU Shi-ming 2, ZHENG Xiu-lian 1   

  1. (1. College of Electrical and Information Engineering, Quzhou University, Quzhou 324000, China; 2. Intelligence Information Systems Research Institute, Zhejiang University of Technology, Hanzhou 310014, China)
  • Received:2011-08-25 Online:2012-03-05 Published:2012-03-05

摘要: 为解决动力学参数估计的问题,提出一种控制参数自适应和策略自适应的差分进化算法(DE-CPASA)。在该算法中,采用差分进化对优化问题进行寻优,控制参数以正态分布的方式实现自适应,通过对适应度函数值的评价,实现变异策略的自适应。测试仿真结果表明,DE-CPASA算法具有较高的求解精度和较快的收敛速度。将DE-CPASA算法应用于Hg氧化动力学参数估计,可得到较好的优化结果。

关键词: 差分进化算法, 自适应, 参数估计, 水银氧化, 策略自适应, 变异因子

Abstract: In order to solve problem of parameter estimation, an Differential Evolution algorithm with Control Parameter Adaptation and Strategy Adaptation(DE-CPASA) is introduced. In DE-CPASA, differential evolution operator is used to search the optimization results of problems, and Gaussian distribution is employed to implement the adaptive control parameters. The strategy adaptation is achieved by evaluation of value of fitness function. Simulation test results show that DE-CPASA can obtain a higher precision solution and has fast convergence. DE-CPASA is employed to estimate the kinetic parameters of Hg oxidation, and a good optimization result is obtained.

Key words: differential evolution algorithm, adaptive, parameter estimation, Hg oxidation, strategy adaptation, mutation factor

中图分类号: