中文摘要：

采用粒子群优化算法，以电压增益、共模抑制比、输入电阻平方根的三者乘积对输出电阻的比作为适应度函数，对差分一共射两级直接耦合电压串联负反馈放大电路中的电阻做自适应优化。结果显示，只要对电路交流指标加以约束，适应度函数值总会减小。当分别对增大电压增益和减小输出电阻进行限制后，电压增益总是尽量小，输出电阻总是尽量大，以使适应度函数在给定约束下取得最大。经EWB软件对优化参数仿真，结果满足线性放大要求。同时说明了可以调整适应度函数形式，找到最佳电路参数，以满足工程上对放大器指标的不同需求。

英文摘要：

The resistance values of two layers directly coupled differential-common emitter voltage series negative feedback amplifying circuit are adaptively optimized by particle swarm optimization. In the method, the voltage gain, common-mode rejection ratio and the square root of the input resistance are multiplied, and the ratio between the product and the output resistance is used as the fitness func- tion. The results indicate that the fitness function value will decrease under the constraint condition that alternating current index is limited. After increasing voltage gain and decreasing the output resistance are restricted, to derive the maximal fitness function value under the given condition, the voltage gain is as small as possible and the output resistance is as large as possible. Simulation results of the parameters with EWB software can meet the requirements of linear amplification. The results also indicate that the optimal parameters of circuit can be found by adjusting the fitness function.