计算机工程与应用 ›› 2015, Vol. 51 ›› Issue (15): 12-16.

• 博士论坛 • 上一篇    下一篇

2-Adic MRA的浮点数编码遗传算法

崔明义   

  1. 1.河南财经政法大学 计算机与信息工程学院,郑州 450046
    2.郑州成功财经学院 信息工程系,郑州 451200
  • 出版日期:2015-08-01 发布日期:2015-08-14

2-Adic MRA based floating point representation Genetic Algorithm

CUI Mingyi   

  1. 1.School of Computer & Information Engineering, Henan University of Finance and Law, Zhengzhou 450046, China
    2.Department of Information Engineering, Zhengzhou Chenggong University of Finance and Economics, Zhengzhou 451200, China
  • Online:2015-08-01 Published:2015-08-14

摘要: 通过2-Adic多分辨率分析,构造正交小波基;证明所构造正交小波用于浮点数编码消噪的正确性;提出用正交小波在浮点数编码遗传算法中进行消噪变异操作,以消除浮点数编码在遗传环境中所产生的噪音对算法性能的影响;构建基于2-Adic多分辨率分析的遗传算法,并进行了实验。仿真实验表明,提出的算法可明显提高浮点数编码遗传算法的收敛速度和精度,具有较高的可靠性。

关键词: 2-Adic多分辨率分析(MRA), 正交小波, 浮点数编码, 消噪变异, 遗传算法

Abstract: An orthonormal wavelet basis is constructed with 2-Adic multiresolution analysis. Validity of the constructed orthonormal wavelet is proven in denoising on Floating Point Representation Genetic Algorithm(FPRGA). Denoising mutation operation with the orthonormal wavelet is proposed in FPRGA. The aim is to remove noises from floating point representation in genetic environment. The genetic algorithm based on 2-Adic multiresolution analysis is structured. The experiment is done. The simulation experiment indicates that the algorithm can improve obviously convergence rate and precision of FPRGA. It has greater reliability than base algorithm.

Key words: 2-Adic Multiresolution Analysis(MRA), orthonormal wavelet, floating point representation, denoising mutation, Genetic Algorithm(GA)