Study on Statistical Out lier Det ection and Labelling

作     者：Pawel D.Domanski Pawe(l) D.Doma(n)ski

作者机构：Institute of Control and Computation EngineeringWarsaw University of TechnologyWarsaw 00-665Poland 

出 版 物：《International Journal of Automation and computing》 (国际自动化与计算杂志（英文版）)

年 卷 期：2020年第17卷第6期

页      面：788-811页

核心收录：

学科分类：07[理学] 0701[理学-数学] 070101[理学-基础数学] 

主　　题：Outlier detection control loop quality statistical analysis robust estimation heavy tails. 

摘      要：Outliers accompany control engineers in their real life *** trial reality is much richer than eleme ntary linear,quadratic,Gaussian *** appear due to various and varying,often unknown,*** meet research interest in statistical and regression analysis and in data *** are a lot of interesting algorithms and approaches to outlier detection,labelling,filtering and finally ***,their impact on control systems has not been found sufficient attention in *** influence is frequently unnoticed,ignored or not *** work focuses on the subject of outlier detection and labelling in the cont ext of control system performance *** ted statistical data-driven approaches are analyzed,as t hey can be easily implemented with limited a priori *** study consists of a simulation study followed by the analysis of real control *** nt generation mechanisms are Simula ted,like overlapping Gaussian processes,symmetric and asymmetric,artificially shifted points and fat-tailed *** observations are confronted with industrial control loops *** work concludes with a practical procedure,which should help practitioners in dealing with outliers in control engineering temporal data.