Kriging models for forecasting crude unit overhead corrosion

Kyungjae Tak, Junghwan Kim, Hweeung Kwon, Jae Hyun Cho, il Moon

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Crude unit overhead corrosion is a major issue in the refinery field. However, the corrosion models in the literature are difficult to apply to real refinery processes due to the characteristics of corrosion. We propose a Kriging model, an advanced statistical tool for geostatistics, to forecast the corrosion rate in a real refinery plant. Instead of spatial coordinates, the proposed model employs the non-spatial coordinates of six key corrosion variables: H2S, Cl, Fe2+, NH3, pH, and flowrate. The Kriging model is compared with two well-known forecasting models, multiple linear regression and an artificial neural network. To overcome the insufficiency of the number of data sets measured in the plant to use the six non-spatial coordinates, the significance probability is applied to reduce the dimensions from six to four. Among all the developed models in this paper, the Kriging model with four corrosion variables showed the best forecasting performance.

Original languageEnglish
Pages (from-to)1999-2006
Number of pages8
JournalKorean Journal of Chemical Engineering
Volume33
Issue number7
DOIs
Publication statusPublished - 2016 Jul 1

Fingerprint

Corrosion
Metal refineries
Corrosion rate
Linear regression
Neural networks

All Science Journal Classification (ASJC) codes

  • Chemistry(all)
  • Chemical Engineering(all)

Cite this

Tak, Kyungjae ; Kim, Junghwan ; Kwon, Hweeung ; Cho, Jae Hyun ; Moon, il. / Kriging models for forecasting crude unit overhead corrosion. In: Korean Journal of Chemical Engineering. 2016 ; Vol. 33, No. 7. pp. 1999-2006.
@article{b3bcc86a918542248099612c36c52d2f,
title = "Kriging models for forecasting crude unit overhead corrosion",
abstract = "Crude unit overhead corrosion is a major issue in the refinery field. However, the corrosion models in the literature are difficult to apply to real refinery processes due to the characteristics of corrosion. We propose a Kriging model, an advanced statistical tool for geostatistics, to forecast the corrosion rate in a real refinery plant. Instead of spatial coordinates, the proposed model employs the non-spatial coordinates of six key corrosion variables: H2S, Cl−, Fe2+, NH3, pH, and flowrate. The Kriging model is compared with two well-known forecasting models, multiple linear regression and an artificial neural network. To overcome the insufficiency of the number of data sets measured in the plant to use the six non-spatial coordinates, the significance probability is applied to reduce the dimensions from six to four. Among all the developed models in this paper, the Kriging model with four corrosion variables showed the best forecasting performance.",
author = "Kyungjae Tak and Junghwan Kim and Hweeung Kwon and Cho, {Jae Hyun} and il Moon",
year = "2016",
month = "7",
day = "1",
doi = "10.1007/s11814-016-0083-9",
language = "English",
volume = "33",
pages = "1999--2006",
journal = "Korean Journal of Chemical Engineering",
issn = "0256-1115",
publisher = "Springer New York",
number = "7",

}

Kriging models for forecasting crude unit overhead corrosion. / Tak, Kyungjae; Kim, Junghwan; Kwon, Hweeung; Cho, Jae Hyun; Moon, il.

In: Korean Journal of Chemical Engineering, Vol. 33, No. 7, 01.07.2016, p. 1999-2006.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Kriging models for forecasting crude unit overhead corrosion

AU - Tak, Kyungjae

AU - Kim, Junghwan

AU - Kwon, Hweeung

AU - Cho, Jae Hyun

AU - Moon, il

PY - 2016/7/1

Y1 - 2016/7/1

N2 - Crude unit overhead corrosion is a major issue in the refinery field. However, the corrosion models in the literature are difficult to apply to real refinery processes due to the characteristics of corrosion. We propose a Kriging model, an advanced statistical tool for geostatistics, to forecast the corrosion rate in a real refinery plant. Instead of spatial coordinates, the proposed model employs the non-spatial coordinates of six key corrosion variables: H2S, Cl−, Fe2+, NH3, pH, and flowrate. The Kriging model is compared with two well-known forecasting models, multiple linear regression and an artificial neural network. To overcome the insufficiency of the number of data sets measured in the plant to use the six non-spatial coordinates, the significance probability is applied to reduce the dimensions from six to four. Among all the developed models in this paper, the Kriging model with four corrosion variables showed the best forecasting performance.

AB - Crude unit overhead corrosion is a major issue in the refinery field. However, the corrosion models in the literature are difficult to apply to real refinery processes due to the characteristics of corrosion. We propose a Kriging model, an advanced statistical tool for geostatistics, to forecast the corrosion rate in a real refinery plant. Instead of spatial coordinates, the proposed model employs the non-spatial coordinates of six key corrosion variables: H2S, Cl−, Fe2+, NH3, pH, and flowrate. The Kriging model is compared with two well-known forecasting models, multiple linear regression and an artificial neural network. To overcome the insufficiency of the number of data sets measured in the plant to use the six non-spatial coordinates, the significance probability is applied to reduce the dimensions from six to four. Among all the developed models in this paper, the Kriging model with four corrosion variables showed the best forecasting performance.

UR - http://www.scopus.com/inward/record.url?scp=84983028215&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84983028215&partnerID=8YFLogxK

U2 - 10.1007/s11814-016-0083-9

DO - 10.1007/s11814-016-0083-9

M3 - Article

AN - SCOPUS:84983028215

VL - 33

SP - 1999

EP - 2006

JO - Korean Journal of Chemical Engineering

JF - Korean Journal of Chemical Engineering

SN - 0256-1115

IS - 7

ER -