Selection of regressors using correlation analysis to design a Virtual Instrument for an SRU of a refinery

Author

Bella, A. Di ; Graziani, S. ; Napoli, G. ; Xibilia, M.G.

Author_Institution

Univ. degli Studi di Catania, Catania

fYear

2007

fDate

27-29 June 2007

Firstpage

1

Lastpage

6

Abstract

In this paper the problem of regressors selection in Virtual Instruments (VI) design is addressed The VI is designed to replace the on line analyzer of a Sulfur Recovery Unit (SRU) of a large refinery located in Sicily during maintenance operations. It is designed by using nonlinear MA models implemented by a MLP neural network. The use of a set of cross-correlation functions, proposed by Billings and Voon to evaluate the performance of nonlinear models is used to select the regressors of the discrete-time NMA model by implementing an automatic regressor selection algorithm. The designed Soft Sensor has been implemented at the refinery to be tested on line.

Keywords

chemical industry; correlation methods; multilayer perceptrons; regression analysis; virtual instrumentation; MLP neural network; automatic regressor selection algorithm; correlation analysis; cross-correlation function; discrete-time NMA model; nonlinear MA model; soft sensor design; sulfur recovery unit; virtual instrument; Algorithm design and analysis; Atmospheric measurements; Costs; Gases; Industrial plants; Instruments; Monitoring; Neural networks; Pollution measurement; Refining; correlation analysis; moving average models; neural modelling; refinery; regressor selection; virtual instruments;

fLanguage

English

Publisher

ieee

Conference_Titel

Control & Automation, 2007. MED '07. Mediterranean Conference on

Conference_Location

Athens

Print_ISBN

978-1-4244-1282-2

Electronic_ISBN

978-1-4244-1282-2

Type

conf

DOI

10.1109/MED.2007.4433715

Filename

4433715