基于混合遗传算法的模拟电路自动设计研究
A Hybrid Genetic Algorithm with Hyper-mutation and Elitist Strategies for Automated Analog Circuit Design
-
摘要: 模拟电路在电子系统设计中具有重要意义.模拟电路设计包括电路拓扑结构设计和元件值设计, 这两个方面对计算机辅助模拟电路设计都很重要.但是, 传统的遗传算法在不断进化电路元件值时效率低.因此, 本文提出一种具有超突变和精英策略的混合遗传算法HME-GA.该方法不仅可以用来设计电路的拓扑结构, 而且可用于设计电路的元件值.实验结果表明, HEM-GA算法优于简单遗传算法GA, 可作为计算机辅助模拟电路设计的有效方法, 在模拟电路自动化设计中具有非常重要的意义和巨大的应用潜力.Abstract: Analog circuits are of great importance in electronic system design.Considering the lack of simple GA on evolving the analog circuit's components values, this paper proposes a hybrid genetic algorithm with hyper-mutation and elitist strategy.Experimental results show that, the proposed algorithm performs much better than simple GA.This paper proposes an efficient way of automated analog circuit design.It is obvious that, evolutionary methods have significant importance and large potential in analog circuit design automation.
下载:
