模型递推辨识算法及应用

研究了具有线性参数的非线性多项式模型的数据阵L－D分解性质，其中D是对角阵，L是带有单位元素的下三角阵．结果表明，通过因子L可以估计出模型的参数，通过因子D可以选择模型中的项；提出了同时进行模型的结构确定和参数估计的递推辨识算法．该算法可用于船舶运动的实时建模．实际应用结果表明，该算法可以有效地辨识多项式等线性参数的非线性模型．

Recursive Identification Algorithm for a Class Model and Application

Corresponding nonlinear polynomial model with linear parameter, properties of the LD factorization of the data matrix are studied in this paper, where D is a diagonal matrix and L is a lower triangular matrix with unit diagonal elements. It will be shown that the parameters of the model can be identified by the L factor while the terms can be selected by the D factor. A recursive algorithm which can select the terms of the model while estimating the parameters, is presented. The algorithm has been used in real modeling of ship movement.The results of applications show that it can efficiently identify the model with linear parameters, such as polynomial function.