Naslov
Soft Sensors Application for the Estimation of Benzene Content in Reformate

Autori
Ujević Andrijić, Željka ; Bolf, Nenad ; Rolich, Tomislav ; Mohler, Ivan

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Izvornik
Danas znanost - sutra industrija : knjiga sažetaka / Jukić, Ante - Zagreb : Osijek : Hrvatsko društvo kemijskih inženjera i tehnologa (HDKI) ; Prehrambeno tehnološki fakultet Sveučilišta Josipa Jurja Strossmayera u Osijeku, 2012, 49-49

ISBN
978-953-6894-46-8

Skup
14. Ružičkini dani

Mjesto i datum
Vukovar, Hrvatska, 13.09.2012. - 15.09.2012

Vrsta sudjelovanja
Poster

Vrsta recenzije
Domaća recenzija

Ključne riječi
Soft sensors; Optimization; Identification; Estimation; Genetic Algorithm

Sažetak
As vehicle emission standards become more stringent there is an increasing need for continuous monitoring of benzene content in gasoline. Since the on-line analyzers are often unavailable, and laboratory analyses are infrequently, dynamic model of the soft sensors for benzene content estimation of light and heavy reformate are developed. Experimental data is acquired from the refinery distributed control system (DCS) and include continuously measured variables and analyzer assays available on-line. Dynamic models based on Finite Impulse Response (FIR) and Output Error (OE) identification methods are presented. To avoid a trial and error procedure, genetic algorithm model order optimizing is proposed. The models are implemented on advanced application module within DCS where the benzene content is calculated based on dynamic polynomial models and stored in process history database. The results show very good matching with experimental data on validation set, thus proving their usefulness for on-line estimation of benzene content in the light reformate. Moreover, since the process analyzer for benzene content is often out of order due to maintenance the model replaces on-line analyzer. Performance is validated comparing the results with laboratory analyzes carried out 2 or 3 times a day.

Izvorni jezik
Engleski

Znanstvena područja
Kemijsko inženjerstvo

POVEZANOST RADA