Robust identification of linear parameter-varying dual-rate system with non-stationary heavy-tailed noise

IF 3.3 2区计算机科学 Q2 AUTOMATION & CONTROL SYSTEMS

Journal of Process Control Pub Date : 2025-07-21 DOI:10.1016/j.jprocont.2025.103500

Xiang Chen , Ke Li , Fei Liu

{"title":"Robust identification of linear parameter-varying dual-rate system with non-stationary heavy-tailed noise","authors":"Xiang Chen , Ke Li , Fei Liu","doi":"10.1016/j.jprocont.2025.103500","DOIUrl":null,"url":null,"abstract":"<div><div>This paper addresses the identification of linear parameter-varying (LPV) dual-rate systems with non-stationary heavy-tailed measurement noise using a variational Bayesian (VB) approach. It provides a comprehensive analysis of dual-rate sampling and noise distribution variations commonly found in system data. To model the outliers, the Student’s <em>t</em> distribution is employed, and a Bernoulli variable is introduced to construct a Gaussian-Student’s <em>t</em> mixture (GTM) distribution that accounts for non-stationary heavy-tailed noise. The GTM distribution is then transformed into a Gaussian hierarchical model to develop a probabilistic representation of the system. Given the unknown process outputs in the regression vector, this study employs a modified Kalman filter for estimation. Based on the obtained estimates and observed data, a prior distribution is defined to establish a Bayesian framework, allowing for iterative parameter estimation via the VB approach. Finally, the effectiveness of this algorithm is validated through a numerical example and a cascaded tank system.</div></div>","PeriodicalId":50079,"journal":{"name":"Journal of Process Control","volume":"153 ","pages":"Article 103500"},"PeriodicalIF":3.3000,"publicationDate":"2025-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Process Control","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0959152425001283","RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"AUTOMATION & CONTROL SYSTEMS","Score":null,"Total":0}

引用次数: 0

Abstract

This paper addresses the identification of linear parameter-varying (LPV) dual-rate systems with non-stationary heavy-tailed measurement noise using a variational Bayesian (VB) approach. It provides a comprehensive analysis of dual-rate sampling and noise distribution variations commonly found in system data. To model the outliers, the Student’s t distribution is employed, and a Bernoulli variable is introduced to construct a Gaussian-Student’s t mixture (GTM) distribution that accounts for non-stationary heavy-tailed noise. The GTM distribution is then transformed into a Gaussian hierarchical model to develop a probabilistic representation of the system. Given the unknown process outputs in the regression vector, this study employs a modified Kalman filter for estimation. Based on the obtained estimates and observed data, a prior distribution is defined to establish a Bayesian framework, allowing for iterative parameter estimation via the VB approach. Finally, the effectiveness of this algorithm is validated through a numerical example and a cascaded tank system.

查看原文本刊更多论文

具有非平稳重尾噪声的线性变参数双速率系统的鲁棒辨识

本文研究了用变分贝叶斯（VB）方法辨识具有非平稳重尾测量噪声的线性变参双速率系统。它提供了对系统数据中常见的双速率采样和噪声分布变化的全面分析。为了对异常值进行建模，采用Student 's t分布，并引入伯努利变量来构建高斯-Student 's t混合（GTM）分布，该分布考虑了非平稳重尾噪声。然后将GTM分布转换为高斯分层模型，以开发系统的概率表示。考虑到回归向量中的未知过程输出，本研究采用改进的卡尔曼滤波进行估计。根据得到的估计和观测数据，定义先验分布，建立贝叶斯框架，通过VB方法进行迭代参数估计。最后，通过一个数值算例和一个叠联罐系统验证了该算法的有效性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Journal of Process Control 工程技术-工程：化工

CiteScore

7.00

自引率

11.90%

发文量

159

审稿时长

74 days

期刊介绍： This international journal covers the application of control theory, operations research, computer science and engineering principles to the solution of process control problems. In addition to the traditional chemical processing and manufacturing applications, the scope of process control problems involves a wide range of applications that includes energy processes, nano-technology, systems biology, bio-medical engineering, pharmaceutical processing technology, energy storage and conversion, smart grid, and data analytics among others. Papers on the theory in these areas will also be accepted provided the theoretical contribution is aimed at the application and the development of process control techniques. Topics covered include: • Control applications• Process monitoring• Plant-wide control• Process control systems• Control techniques and algorithms• Process modelling and simulation• Design methods Advanced design methods exclude well established and widely studied traditional design techniques such as PID tuning and its many variants. Applications in fields such as control of automotive engines, machinery and robotics are not deemed suitable unless a clear motivation for the relevance to process control is provided.