首页 | 官方网站   微博 | 高级检索  
     

基于Memetic算法的电路演化设计研究
引用本文:莫宏伟,徐立芳.基于Memetic算法的电路演化设计研究[J].电子学报,2013,41(5):1036-1040.
作者姓名:莫宏伟  徐立芳
作者单位:1. 哈尔滨工程大学自动化学院,黑龙江哈尔滨,150001
2. 哈尔滨工程大学工程训练中心,黑龙江哈尔滨,150001
基金项目:国家自然科学基金,黑龙江省杰出青年基金,中央高校基本科研业务重大专项
摘    要: 针对传统演化算法在设计数字逻辑电路时存在的演化速度缓慢和容易陷入局部最优解等问题,设计了一种Cartesian进化编程编码的电路演化Memetic算法,采用遗传算法作为全局搜索方法,并设计了适合电路演化的基本门种类局部搜索策略.通过一位全加器电路证明所提出的Memetic硬件演化算法的搜索能力,实验证明所提出的算法能够有效地完成进化任务,具有较强的全局快速和局部搜索能力.

关 键 词:演化硬件  演化算法  Memetic算法  局部搜索策略
收稿时间:2012-04-19

Research on Evolvable Hardware Design Based on Memetic Algorithm
MO Hong-wei , XU Li-fang.Research on Evolvable Hardware Design Based on Memetic Algorithm[J].Acta Electronica Sinica,2013,41(5):1036-1040.
Authors:MO Hong-wei  XU Li-fang
Affiliation:1. Automation College,Harbin Engineering University,Harbin,Heilongjiang 150001,China;
2. Engineering Training Center,Harbin Engineering University,Harbin,Heilongjiang 150001,China
Abstract:In order to overcome the problems of traditional evolution algorithms in designing digital logic circuits,such as slow evolution speed and premature convergence,a memetic algorithm (MA) for evolving digital circuits is presented based on CGP encoding.It adopts genetic algorithm as global search method.A local search strategy based on gate type is proposed.The search ability of the proposed method is tested by evolving one-bit full adder.Experimental results show that it is efficient in solving the problems of hardware evolution and has the ability of fast global and local convergence.
Keywords:evolvable hardware  evolutionary algorithm  memetic algorithm  local search strategy
本文献已被 万方数据 等数据库收录!
点击此处可从《电子学报》浏览原始摘要信息
点击此处可从《电子学报》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司    京ICP备09084417号-23

京公网安备 11010802026262号