提出一种新的基因操作策略 ,该策略利用单纯形法的思想产生新样本 ,将遗传算法寻优的随机性与传统算法寻优的方向性有机地结合在一起。仿真结果表明 ,将改进的遗传算法用于训练神经网络辨识器 ,可提高收敛速度和模型拟合精度 A new gene manipulation strategy was proposed. This strategy used the idea of ​​simplex method to generate a new sample, and combined the randomness of optimization of genetic algorithm with the direction of optimization of traditional algorithm. The simulation results show that the improved genetic algorithm can be used to train the neural network identifier to improve the convergence speed and model fitting accuracy