Soft sensor for NOx and O2 using dynamic neural networks

M. Shakil; M. Elshafei; M. A. Habib; F. A. Maleki

doi:10.1016/j.compeleceng.2008.08.007

Soft sensor for NO_x and O₂ using dynamic neural networks

M. Shakil^*
, M. Elshafei
, M. A. Habib
, F. A. Maleki

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

99 Scopus citations

Abstract

Inferential or soft sensing techniques have been gaining momentum recently as viable alternatives to hardware sensors in various situations, e.g. continuous emission monitoring systems. Dynamic neural networks are used in the present work to develop soft sensors for the NO_x and O₂ emission due to combustion operation in industrial boilers. A simplified structure for the soft sensor is obtained by grouping the input variables, reducing the input data dimension and utilizing the system knowledge. The principal component analysis (PCA) is used to reduce the input data dimension. The genetic algorithm (GA) is used to estimate the system's time delays by optimizing a linear time-delay model. Real data from a boiler plant is used to validate the models. The performance of the proposed dynamic models is compared with static neural network models. The results demonstrate the effectiveness of the proposed models.

Original language	English
Pages (from-to)	578-586
Number of pages	9
Journal	Computers and Electrical Engineering
Volume	35
Issue number	4
DOIs	https://doi.org/10.1016/j.compeleceng.2008.08.007
State	Published - Jul 2009

Bibliographical note

Funding Information:
We acknowledge the help of the Saudi Petrochemical Co (Sadaf) for providing the necessary data and technical information. The authors would also like to acknowledge the support of King Fahd University of Petroleum and Minerals. This work was initiated by the KFUPM/SABIC Project No. 2004/06.

Keywords

Artificial neural network (ANN)
Boiler control
Inferential sensing
NO emission
Predictive emission monitoring (PEM)
Soft sensor

ASJC Scopus subject areas

Control and Systems Engineering
General Computer Science
Electrical and Electronic Engineering

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1016/j.compeleceng.2008.08.007

Cite this

@article{4b12849853714259af84169e6d344fc7,

title = "Soft sensor for NOx and O2 using dynamic neural networks",

abstract = "Inferential or soft sensing techniques have been gaining momentum recently as viable alternatives to hardware sensors in various situations, e.g. continuous emission monitoring systems. Dynamic neural networks are used in the present work to develop soft sensors for the NOx and O2 emission due to combustion operation in industrial boilers. A simplified structure for the soft sensor is obtained by grouping the input variables, reducing the input data dimension and utilizing the system knowledge. The principal component analysis (PCA) is used to reduce the input data dimension. The genetic algorithm (GA) is used to estimate the system's time delays by optimizing a linear time-delay model. Real data from a boiler plant is used to validate the models. The performance of the proposed dynamic models is compared with static neural network models. The results demonstrate the effectiveness of the proposed models.",

keywords = "Artificial neural network (ANN), Boiler control, Inferential sensing, NO emission, Predictive emission monitoring (PEM), Soft sensor",

author = "M. Shakil and M. Elshafei and Habib, \{M. A.\} and Maleki, \{F. A.\}",

note = "Funding Information: We acknowledge the help of the Saudi Petrochemical Co (Sadaf) for providing the necessary data and technical information. The authors would also like to acknowledge the support of King Fahd University of Petroleum and Minerals. This work was initiated by the KFUPM/SABIC Project No. 2004/06. ",

year = "2009",

month = jul,

doi = "10.1016/j.compeleceng.2008.08.007",

language = "English",

volume = "35",

pages = "578--586",

journal = "Computers and Electrical Engineering",

issn = "0045-7906",

number = "4",

}

TY - JOUR

T1 - Soft sensor for NOx and O2 using dynamic neural networks

AU - Shakil, M.

AU - Elshafei, M.

AU - Habib, M. A.

AU - Maleki, F. A.

N1 - Funding Information: We acknowledge the help of the Saudi Petrochemical Co (Sadaf) for providing the necessary data and technical information. The authors would also like to acknowledge the support of King Fahd University of Petroleum and Minerals. This work was initiated by the KFUPM/SABIC Project No. 2004/06.

PY - 2009/7

Y1 - 2009/7

N2 - Inferential or soft sensing techniques have been gaining momentum recently as viable alternatives to hardware sensors in various situations, e.g. continuous emission monitoring systems. Dynamic neural networks are used in the present work to develop soft sensors for the NOx and O2 emission due to combustion operation in industrial boilers. A simplified structure for the soft sensor is obtained by grouping the input variables, reducing the input data dimension and utilizing the system knowledge. The principal component analysis (PCA) is used to reduce the input data dimension. The genetic algorithm (GA) is used to estimate the system's time delays by optimizing a linear time-delay model. Real data from a boiler plant is used to validate the models. The performance of the proposed dynamic models is compared with static neural network models. The results demonstrate the effectiveness of the proposed models.

AB - Inferential or soft sensing techniques have been gaining momentum recently as viable alternatives to hardware sensors in various situations, e.g. continuous emission monitoring systems. Dynamic neural networks are used in the present work to develop soft sensors for the NOx and O2 emission due to combustion operation in industrial boilers. A simplified structure for the soft sensor is obtained by grouping the input variables, reducing the input data dimension and utilizing the system knowledge. The principal component analysis (PCA) is used to reduce the input data dimension. The genetic algorithm (GA) is used to estimate the system's time delays by optimizing a linear time-delay model. Real data from a boiler plant is used to validate the models. The performance of the proposed dynamic models is compared with static neural network models. The results demonstrate the effectiveness of the proposed models.

KW - Artificial neural network (ANN)

KW - Boiler control

KW - Inferential sensing

KW - NO emission

KW - Predictive emission monitoring (PEM)

KW - Soft sensor

UR - https://www.scopus.com/pages/publications/67349256117

U2 - 10.1016/j.compeleceng.2008.08.007

DO - 10.1016/j.compeleceng.2008.08.007

M3 - Article

AN - SCOPUS:67349256117

SN - 0045-7906

VL - 35

SP - 578

EP - 586

JO - Computers and Electrical Engineering

JF - Computers and Electrical Engineering

IS - 4

ER -