针对传递函数的模型降阶工作 ,提出采样点在对数分度采样区间上均匀分布的频域采样方案 ,以及基于对数的最小二乘指标。应用遗传算法对此复杂的目标函数进行了求解 ,并通过与传统方法的比较 ,说明了遗传算法在此方面应用的优越性 In order to reduce the work of the transfer function model, a frequency domain sampling scheme with uniform distribution of sampling points on logarithmic sampling intervals and a least-square index based on logarithm are proposed. The complex objective function is solved by genetic algorithm, and the superiority of genetic algorithm in this aspect is illustrated through the comparison with the traditional method