基于复值稀疏Bayesian的系统稳定性辨识
System Stability Identification Based on Complex-Valued Sparse Bayesian
谢伟翔 1陈安琪1
作者信息
- 1. 韶关学院 数学与统计学院,广东 韶关 512005
- 折叠
摘要
稀疏Bayesian学习是近年来机器学习研究的热点,基于Szegö核的复值稀疏Bayesian学习算法能提供稀疏的有理逼近.提出基于Szegö核的复值稀疏Bayesian学习算法来判定单位圆盘内闭环系统的稳定性,该方法具有可给出逼近的解析表达式和适用范围更广的优点,并且不需要参数控制进行迭代优化,运算速度快.实验结果表明,此方法是有效的.
Abstract
Sparse Bayesian learning is a hot topic in machine learning research in recent years,and the complex-valued sparse Bayesian learning algorithm based on the Szegö kernel can provide sparse rational approximation.A complex-valued sparse Bayesian learning algorithm based on the Szegö ̈kernel is proposed to determine the stability of the closed-loop system in the unit disk.This method has the advantages of giving approximate analytical expressions and a wider range of applications,and does not require parameter control for iteration optimized for fast operation.Experimental results show that this method is effective.
关键词
稀疏Bayesian/稳定系统/Szegö核/稳定性判据Key words
sparse Bayesian/stable system/Szegö kernel/stability criterion引用本文复制引用
基金项目
韶关学院校级自然科学类科研项目(SY2021KJ11)
出版年
2024
NETL
NSTL
万方数据