In this paper, the true values of the parameters of the difference equation equivalent to a class of nonlinear system state space equations are derived. Four commonly used parameter identification algorithms are used to identify the parameters of the difference equation, and the accuracy of the algorithms is compared. The parameters of the equivalent difference equation for a nonlinear system are identified by using the forgetting factor recurrent least squares (FFRLS), recurrent gradient correction (RGC), recurrent stochastic Newton algorithm (RSNA) and back propagation neural network(BP). By comparing the identification results with the real value, it is found that the least square method is the most accurate parameter identification algorithm.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Characteristic Modeling a Class of Nonliear Systems with Different Parameter Estimation Methods

  • Yiyang Zeng,
  • Haoshuai Wang,
  • Lei Chen,
  • Zhaoqi Dong

摘要

In this paper, the true values of the parameters of the difference equation equivalent to a class of nonlinear system state space equations are derived. Four commonly used parameter identification algorithms are used to identify the parameters of the difference equation, and the accuracy of the algorithms is compared. The parameters of the equivalent difference equation for a nonlinear system are identified by using the forgetting factor recurrent least squares (FFRLS), recurrent gradient correction (RGC), recurrent stochastic Newton algorithm (RSNA) and back propagation neural network(BP). By comparing the identification results with the real value, it is found that the least square method is the most accurate parameter identification algorithm.