...
首页> 外文期刊>Machine Learning Research >Recursive Algorithms of Closed Loop Identification with a Tailor Made Parameterization
【24h】

Recursive Algorithms of Closed Loop Identification with a Tailor Made Parameterization

机译:量身定制的参数化闭环辨识的递归算法

获取原文

摘要

In this paper, we propose two recursive algorithms for closed loop identification under the framework of a tailor made parameterization. The closed loop transfer function is parameterized using the parameters of the open loop plant model, and utilizing knowledge of the feedback controller. When the plant model and feedback controller are all polynomial forms, a recursive least squares method with forgetting schemes is proposed to verify that this recursive method can be regarded as regularization least squares problem. Furthermore we also extend the tailor made parameterization method to nonlinear system and nonlinear controller, then an iterative least squares algorithm is applied to solve one nonlinear optimization problem.
机译:在本文中,我们提出了两种在定制参数化框架下用于闭环识别的递归算法。使用开环工厂模型的参数并利用反馈控制器的知识对闭环传递函数进行参数化。当工厂模型和反馈控制器均为多项式形式时,提出了一种具有遗忘方案的递推最小二乘法,以验证该递归方法可以看作是正则化最小二乘问题。此外,我们还将量身定制的参数化方法扩展到非线性系统和非线性控制器,然后应用迭代最小二乘算法解决一个非线性优化问题。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号