Investigating effects of pre-processing variation on time-varying steam temperature of distillation column

The system identification involves the preprocessing technique which separates the dataset into two parts; training set and testing set to estimate and validate the model respectively. The main aim of this study is to investigate the effects of pre-processing techniques on the steam temperature of t...

Full description

Bibliographic Details
Published in:Proceeding - 2019 IEEE 7th Conference on Systems, Process and Control, ICSPC 2019
Main Author: Hambali N.; Taib M.N.; Yassin A.I.M.; Rahiman M.H.F.
Format: Conference paper
Language:English
Published: Institute of Electrical and Electronics Engineers Inc. 2019
Online Access:https://www.scopus.com/inward/record.uri?eid=2-s2.0-85084839255&doi=10.1109%2fICSPC47137.2019.9068058&partnerID=40&md5=769cc83b59a53424772f237584bd8c2c
Description
Summary:The system identification involves the preprocessing technique which separates the dataset into two parts; training set and testing set to estimate and validate the model respectively. The main aim of this study is to investigate the effects of pre-processing techniques on the steam temperature of the distillation column. Magnitude scaling, block division and interleaving methods were implemented for steam temperature. Two perturbation input signals, of Pseudo Random Binary Signal (PRBS) and Multi-level Pseudo Random Sequence (MPRS) were employed for Steam Distillation Pilot Plant's (SDPP). The analysis was made based on the fitness and Correlation Violations (CRV) values from the combinations of pre-processing methods. The results of this study indicated that the interleaving technique presented a significant good reduction of CRV for PRBS (39 to 47 CRVs) and MPRS (23 to 30 CRVs) together with low fitness for all the criterion models. © 2019 IEEE.
ISSN:
DOI:10.1109/ICSPC47137.2019.9068058