Naslov
Prediction of sulphur content in the industrial hydrotreatment process

Autori
Lukec, Ivana ; Sertić-Bionda, Katica ; Lukec, Darko

Izvornik
Fuel Processing Technology (0378-3820) 89 (2008), 3; 292-300

Vrsta, podvrsta i kategorija rada
Radovi u časopisima, članak, znanstveni

Ključne riječi
Soft sensor; Hydrotreatment; Neural networks; Quality control

Sažetak
The artificial neural network models were developed to determine sulphur content in the hydrotreatment product. Two models for two different types of feed were developed: light gas oil and vacuum gas oil. The developed ANN models use 6 input variables that are continuously measured in the process and are in accordance with the engineering knowledge and thermodynamics of hydrotreatment processes. Given models show good predictability of sulphur content in the hydrotreatment product and are, therefore, used in practice for continuous monitoring and optimization. This kind of application can be easily developed in any other hydrotreatment process with the available adequate historical data.

Izvorni jezik
Engleski

Znanstvena područja
Kemijsko inženjerstvo

POVEZANOST RADA