Poisoning fault diagnosis in chemical gas sensor arrays using multivariate statistical signal processing and structured residuals generation

M. Padilla; A. Pererat; I. Montoliut; A. Chaudry; K. Persaud; S. Marcot

doi:10.1109/WISP.2007.4447638

Poisoning fault diagnosis in chemical gas sensor arrays using multivariate statistical signal processing and structured residuals generation

M. Padilla, A. Pererat, I. Montoliut, A. Chaudry, K. Persaud, S. Marcot

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

Abstract

Chemical gas sensors are a cheaper and faster alternative for gas analysis than conventional analytic instruments. .However they are prone to degradation because of sensor poisoning and drift. Statistical methods like Principal Component Analysis (PCA) and Partial Least Squares (PLS) have been proved to be very useful in the task of fault diagnosis of malfunctioning sensors. In this work we test the effectiveness of several techniques based on PCA and PLS on faults caused by sensor poisoning. These techniques will be evaluated on a dataset composed by the signals of 17 conductive polymers gas sensors measuring three analytes at several concentration levels. These techniques will be evaluated concerning their capabilities to detect the fault, identify the faulty sensor and correct their signal. ©2007 IEEE.

Original language	English
Title of host publication	2007 IEEE International Symposium on Intelligent Signal Processing, WISP
DOIs	https://doi.org/10.1109/WISP.2007.4447638
Publication status	Published - 2007

Keywords

Fault diagnosis
Gas sensor array
Poisoning

Access to Document

10.1109/WISP.2007.4447638

Cite this

@inproceedings{9fe5c0e395b7421791204222b38f5239,

title = "Poisoning fault diagnosis in chemical gas sensor arrays using multivariate statistical signal processing and structured residuals generation",

abstract = "Chemical gas sensors are a cheaper and faster alternative for gas analysis than conventional analytic instruments. .However they are prone to degradation because of sensor poisoning and drift. Statistical methods like Principal Component Analysis (PCA) and Partial Least Squares (PLS) have been proved to be very useful in the task of fault diagnosis of malfunctioning sensors. In this work we test the effectiveness of several techniques based on PCA and PLS on faults caused by sensor poisoning. These techniques will be evaluated on a dataset composed by the signals of 17 conductive polymers gas sensors measuring three analytes at several concentration levels. These techniques will be evaluated concerning their capabilities to detect the fault, identify the faulty sensor and correct their signal. {\textcopyright}2007 IEEE.",

keywords = "Fault diagnosis, Gas sensor array, Poisoning",

author = "M. Padilla and A. Pererat and I. Montoliut and A. Chaudry and K. Persaud and S. Marcot",

note = "Times Cited: 0 Urena, JU Dominguez, JJG IEEE International Symposium on Intelligent Signal Processing OCT 03-05, 2007 Alcala de Henares, SPAIN",

year = "2007",

doi = "10.1109/WISP.2007.4447638",

language = "English",

booktitle = "2007 IEEE International Symposium on Intelligent Signal Processing, WISP",

}

TY - GEN

T1 - Poisoning fault diagnosis in chemical gas sensor arrays using multivariate statistical signal processing and structured residuals generation

AU - Padilla, M.

AU - Pererat, A.

AU - Montoliut, I.

AU - Chaudry, A.

AU - Persaud, K.

AU - Marcot, S.

N1 - Times Cited: 0 Urena, JU Dominguez, JJG IEEE International Symposium on Intelligent Signal Processing OCT 03-05, 2007 Alcala de Henares, SPAIN

PY - 2007

Y1 - 2007

N2 - Chemical gas sensors are a cheaper and faster alternative for gas analysis than conventional analytic instruments. .However they are prone to degradation because of sensor poisoning and drift. Statistical methods like Principal Component Analysis (PCA) and Partial Least Squares (PLS) have been proved to be very useful in the task of fault diagnosis of malfunctioning sensors. In this work we test the effectiveness of several techniques based on PCA and PLS on faults caused by sensor poisoning. These techniques will be evaluated on a dataset composed by the signals of 17 conductive polymers gas sensors measuring three analytes at several concentration levels. These techniques will be evaluated concerning their capabilities to detect the fault, identify the faulty sensor and correct their signal. ©2007 IEEE.

AB - Chemical gas sensors are a cheaper and faster alternative for gas analysis than conventional analytic instruments. .However they are prone to degradation because of sensor poisoning and drift. Statistical methods like Principal Component Analysis (PCA) and Partial Least Squares (PLS) have been proved to be very useful in the task of fault diagnosis of malfunctioning sensors. In this work we test the effectiveness of several techniques based on PCA and PLS on faults caused by sensor poisoning. These techniques will be evaluated on a dataset composed by the signals of 17 conductive polymers gas sensors measuring three analytes at several concentration levels. These techniques will be evaluated concerning their capabilities to detect the fault, identify the faulty sensor and correct their signal. ©2007 IEEE.

KW - Fault diagnosis

KW - Gas sensor array

KW - Poisoning

U2 - 10.1109/WISP.2007.4447638

DO - 10.1109/WISP.2007.4447638

M3 - Conference contribution

BT - 2007 IEEE International Symposium on Intelligent Signal Processing, WISP

ER -

Poisoning fault diagnosis in chemical gas sensor arrays using multivariate statistical signal processing and structured residuals generation

Abstract

Keywords

Access to Document

Fingerprint

Cite this