基于遗传算法的抽指加权叉指换能器进化设计

提出了极性抽指加权叉指换能器设计的新方法，克服了传统设计方法的繁杂性。将极性抽指加权叉指换能器作为染色体，通过独特的（-1，1）的二值编码，以目标频率响应曲线和待进化的叉指换能器频率响应曲线在考虑的频率范围内的1601个采样点的误差值为进化目标，对种群中的染色体进行选择、交叉和变异等遗传操作，自动进化出符合目标要求的极性抽指加权叉指换能器极性加权状况。进化实验结果表明，应用本文提出的进化方法设计出的极性抽指加权叉指换能器的频响曲线与目标频响曲线基本重合，达到设计要求，进化设计方法效率高，实用性强。

Evolutionary Design for Withdrawal Weighting Transducer Based on Genetic Algorithm

A new method to design source-withdrawal weighting transducer is presented,which avoids the complexity when using traditional design methods.Source-withdrawal weighting transducer is regarded as chromosome with(-1,1) two-value code and error between target frequency response and frequency response of evoluting source-withdrawal weighting transducer at 1601 frequency value as evolutionary object.Source-withdrawal weighting transducer,which satisfies target frequency response,is obtained after selection,crossover and mutation operation.Evolutionary experiment shows that the frequency response of source-withdrawal weighting transducer obtained by evolution is in agreement with target frequency response.The evolutionary design method is efficient and practical.