快速实现数字仿生电路设计的自适应遗传算法

Self-adaptive Genetic Algorithm to Design Efficiently Digital Evolvable Hardware

采用演化硬件技术快速实现数字仿生电路的设计是演化硬件一个重要的研究方向;对演化算法的改进是提高演化速率和减少计算机计算负荷的重要方法;借鉴演化策略和模拟退火算法的思想以及Levi提出的HereBoy算法,提出了具有自适应能力的增强型演化算法;通过实验发现,该算法在演化相同的数字逻辑电路中,明显地提高了演化的速度,缩短了演化时间,提高了进化设计的速度、规模和优化程度。

To implement the digital bionic circuit based on evolvable hardware is one of important research area. The important approach to improve the speed of evolution and shorten the time of evolution is to improve the genetic algorithm. Based on HereBoy algorithm, the en-hanced genetic algorithm which references on evolution strategy （ES） and Simulated Annealing is proposed. The experiment results show that the speed of evolution is accelerated and the time of evolution is shorten. Then, the scale and structure of circuit are improved.