摘要
为提高演化硬件在演化过程中的收敛速度,实现复杂的演化硬件,研究以Xilinx公司的Virtex-5Pro系列开发板作为硬件平台的基于SOPC的自演化系统。分析简单遗传算法与量子遗传算法对种群的适应度以及收敛速度的影响;实验中通过全加器电路和2位乘法器电路实现了自演化系统的验证。结合实例,对2种算法分别进行仿真,仿真结果表明,相对于标准遗传算法而言,量子遗传算法效率更高、更适应于进化复杂的大规模电路。
To improve the convergence of the evolvable hardware in the process of evolution and to realize the complex evolvable hardware, the evolvable hardware system which took Virtex-5 Pro FPGA development board as the hardware platform was studied. At the same time, the effects of the standard genetic algorithm and the quantum genetic algorithm on the convergence rate as well as fitness were analyzed. An adder circuit and a two-bit multiplier circuit were used to verify the evolvable hardware system. The two algorithms were emulated. It is proved that compared to the standard genetic algorithm, the quantum genetic algorithm is more suitable for the evolution of the complex large-scale circuit and the convergence rate is indeed improved.
出处
《计算机工程与设计》
CSCD
北大核心
2014年第9期3244-3248,共5页
Computer Engineering and Design
基金
国家自然科学基金项目(60871009)
南京航空航天大学基本科研业务费专项科研基金项目(XNA201288)
关键词
演化硬件
简单遗传算法
量子遗传算
适应度
收敛性
evolvable hardware
standard genetic algorithm
quantum genetic algorithm
fitness
convergence
