Premium
Comparison of methods for estimation of the covariance matrix of measurement errors
Author(s) -
Feital Thiago,
Prata Diego Martinez,
Pinto José Carlos
Publication year - 2014
Publication title -
the canadian journal of chemical engineering
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.404
H-Index - 67
eISSN - 1939-019X
pISSN - 0008-4034
DOI - 10.1002/cjce.22063
Subject(s) - observational error , process (computing) , computer science , propagation of uncertainty , measurement uncertainty , covariance matrix , covariance , algorithm , accuracy and precision , work in process , process control , data mining , mathematics , statistics , engineering , operations management , operating system
Measurement errors normally present in measured process variables can mask the actual process conditions and lead to wrong operational decisions. This is due not only to the imprecise measured values but also to the propagation of the measurement error through process models used to perform control and optimization tasks. One possible alternative to deal with this problem is to characterize the measurement errors online and to incorporate the obtained information in the process operation. The present work analyzes the performances of different methods used for characterization of measurement errors in different scenarios commonly found in the industrial practice. Based on the obtained results, a method is proposed for direct evaluation of measurement errors at plant site.