Original scientific paper

SOFT SENSOR FOR DEBUTANIZER COLUMN PROCESS CONTROL

Ivica Jerbić
Hrvoje Pavelić
Nenad Bolf

Abstract

Abstract
Law regulations dictate restrictions of product quality specifications and refinery emissions. Measurement of great number of process variables and installing new expensive process analyzers is necessary for efficient process control. Possible solution of this problem is application of soft-sensors i.e. the process model that is used for variable estimations which are not measured in continuous manner.
This paper describes soft-sensor design for product quality monitoring and process control of debutanizer column of INA Oil Refinery Sisak. Neural network model is developed based on available process measurement with the purpose of product quality monitoring and process control enhancement.
Method of estimation of pentane fraction in liquefied petroleum gas (LPG) and Reid vapour pressure of stabilized FCC gasoline using inferential model is elaborated. Soft- sensor deals with lag of laboratory analysis and provides on-line product quality monitoring. The aim is to control process in debutanizer column thus pentane fraction in LPG is kept under 2 mass percent and RVP of FCC gasoline on desired value (50 kPa). The investigation involves the simulation study and validation of achieved results by comparison with the experimental data.

Keywords

neural network; selflearning control system; selforganizing control system

Hrčak ID:

12308

URI

https://hrcak.srce.hr/12308

Publication date:

2.5.2007.

Article data in other languages: croatian

Visits: 3.059 *