有理模型辨识的两类新方法—混合迭代与柔性最小二乘法
作者:
作者单位:

1.江南大学;2.西英格兰大学

作者简介:

通讯作者:

中图分类号:

TP273

基金项目:

国家自然科学基金项目(面上项目,重点项目,重大项目)


Two novel identification methods for rational models—compound iterative algorithm and flexible least squares algorithm
Author:
Affiliation:

1.Jiangnan University;2.University of the West of England

Fund Project:

This work is supported by the National Natural Science Foundation of China (No.61973137)

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    针对有理模型提出了两类辨识方法. 首先提出基于递阶辨识思想的混合辨识方法, 将模型分解为分子和分母两个子模型, 分别用最小二乘法辨识分子参数, 用粒子群算法和智能多步长梯度迭代算法辨识分母参数. 由于降低了模型维数, 且信息向量和噪声不相关, 因此相对于传统的偏差补偿最小二乘算法, 混合迭代法提高了辨识精度并降低了计算量. 为消除模型结构已知假设, 且充分利用最新数据更新系统参数, 提出了柔性递推最小二乘辨识方法, 将有理模型转化为时变参数系统, 进而辨识出时变系统的参数. 仿真例子验证了本文方法的有效性.

    Abstract:

    This study proposes two identification methods for nonlinear rational models. The first is the compound iterative algorithm which is based on the hierarchical technique. Compared with the traditional identification methods, the compound iterative algorithm transforms the rational model into two sub-models whose parameters are estimated iteratively, thus it has less computational efforts and higher estimation accuracy. In addition, a flexible recursive least squares algorithm is proposed, which does not require the knowledge of the structure of the denominator model, and can update the parameters with new collected data. The simulation results verify the effectiveness of the algorithms.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2020-07-10
  • 最后修改日期:2021-09-23
  • 录用日期:2020-11-05
  • 在线发布日期: 2020-12-01
  • 出版日期: