欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
考虑有色噪声干扰的Hammerstein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。
英文摘要：
The identification problems of Hammerstein nonlinear systems with colored noise are considered. An extended projection algorithm, a simplified extended projection algorithm and an extended stochastic gradient （ESG） identification algorithm are presented by using the gradient search principle. The basic idea is to replace the unmeasurable noise terms in the extended information vectors with their estimates. Since the extended projection algorithm is sensitive to noise and the ESG algorithm has a slow convergence rate, a forgetting factor is introduced in the algorithms to improve the convergence rate. Furthermore, the performances of these approaches are analyzed and compared by using a numerical example.
同期刊论文项目
同项目期刊论文
Auxiliary model based least squares identification method for a state space model with a unit time-d
Recursive Relations of the Cost Functions for the Least-Squares Algorithms for Multivariable Systems
Hierarchical least-squares based iterative identification for multivariable systems with moving aver
Maximum likelihood least squares identification for systems with autoregressive moving average noise
An iterative least squares estimation algorithm for controlled moving average systems based on matri
Least squares estimation for a class of non-uniformly sampled systems based on the hierarchical iden
Hierarchical multi-innovation stochastic gradient algorithm for Hammerstein nonlinear system modelin
期刊信息
位置：