| 引用本文: | 柴秀俊,王宏伟,王林,嵇薪儒.基于高斯混合聚类的切换系统的辨识[J].控制理论与应用,2021,38(5):634~640.[点击复制] |
| CHAI Xiu-jun,WANG Hong-wei,WANG Lin,JI Xin-ru.Identification of switched systems based on Gaussian mixture clustering[J].Control Theory and Technology,2021,38(5):634~640.[点击复制] |
|
| 基于高斯混合聚类的切换系统的辨识 |
| Identification of switched systems based on Gaussian mixture clustering |
| 摘要点击 1614 全文点击 605 投稿时间：2020-04-12 修订日期：2020-10-07 |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| DOI编号 10.7641/CTA.2020.00194 |
| 2021,38(5):634-640 |
| 中文关键词 切换系统 模式检测 高斯混合聚类 递推增广最小二乘法 贝叶斯信息准则 |
| 英文关键词 switched systems mode detection Gaussian mixture clustering recursive extended least square method Bayesian information criterion |
| 基金项目 国家自然科学基金项目(61863034)资助. |
|
| 中文摘要 |
| 针对具有未知切换规则与未知子系统数量的切换系统的辨识问题, 提出一种两阶段辨识方法, 包括模式检
测与参数辨识. 在模式检测阶段, 首先建立高斯混合模型表示采样数据的分布, 并通过轮盘法选择合适的初始模型
参数. 其次, 计算采样数据属于每个子系统的后验概率, 通过极大似然估计算法迭代更新模型参数, 使高斯混合模
型最大化地拟合采样数据的分布. 在此基础上, 通过贝叶斯信息准则确定子系统的数量, 并根据最大后验概率准则
估计切换规则. 在参数辨识阶段, 通过递推增广最小二乘法估计每个子系统的参数向量. 最后, 通过仿真结果验证
了所提方法的有效性. |
| 英文摘要 |
| In order to solve the identification problem of switched systems with unknown switched rules and unknown
number of subsystems, a two-stage identification method is proposed, including mode detection and parameter identification.
In the mode detection stage, the Gaussian mixture model is first established to represent the distribution of sampled
data, and appropriate initial model parameters are selected according to the roulette method. Secondly, the posterior probability
of the sampled data belong to each subsystem is calculated, and the maximum likelihood estimation algorithm is used
to iteratively update the model parameters to make the Gaussian mixture model maximum fit the distribution of the sampled
data. On this basis, the number of subsystems is determined by the Bayesian information criterion, and the switched rule
is estimated according to the maximum a posteriori criterion. In the parameter identification stage, the parameter vector
of each subsystem is estimated by the recursive extended least square method. Finally, the effectiveness of the proposed
method is verified according to simulation results. |
|
|
|
|
|