考虑有色噪声干扰的Hamm erste in非线性系统的辨识,通过梯度搜索原理推导了增广投影算法,简化增广投影算法和增广随机梯度辨识算法。基本思想是将增广信息向量中的未知噪声项用其估计残差代替。增广投影算法对噪声非常敏感,增广随机梯度算法的收敛速度慢,为了解决这些不足,在增广随机梯度算法中引入遗忘因子,来改善参数估计精度,进一步通过仿真来比较算法的估计误差以及收敛速度。 Considering the recognition of Hammesste in non-linear system with colored noise, the augmented projection algorithm, the simplified augmented projection algorithm and the augmented stochastic gradient identification algorithm are derived through the principle of gradient search. The basic idea is to replace the unknown noise term in augmented information vector with its estimated residuals. The augmented projection algorithm is very sensitive to noise. The augmented stochastic gradient algorithm has a slow convergence rate. To solve these problems, the forgetting factor is introduced into the augmented stochastic gradient algorithm to improve the accuracy of the parameter estimation. The estimation error of the algorithm is further compared by simulation As well as the convergence speed.