TY - BOOK
T1 - Fault detection, identification, and reconstruction of faulty chemical gas sensors under drift conditions, using Principal Component Analysis and Multiscale-PCA
AU - Padilla, M.
AU - Perera, A.
AU - Montoliu, I.
AU - Chaudry, A.
AU - Persaud, K.
AU - Marco, S.
N1 - S20100
PY - 2010
Y1 - 2010
N2 - Statistical methods like Principal Components Analysis (PCA) or Partial Least Squares (PLS) and multiscale approaches, have been reported to be very useful in the task of fault diagnosis of malfunctioning sensors for several types of faults. In this work, we compare the performance of PCA and Multiscale-PCA on a fault based on a change of sensor sensitivity. This type of fault affects chemical gas sensors and it is one of the effects of the sensor poisoning. These two methods will be applied on a dataset composed by the signals of 17 conductive polymer gas sensors, measuring three analytes at several concentration levels during 10 months. Therefore, additionally to performance's comparison, both method's stability along the time will be tested. The comparison between both techniques will be made regarding three aspects; detection, identification of the faulty sensors and correction of faulty sensors response. © 2010 IEEE.
AB - Statistical methods like Principal Components Analysis (PCA) or Partial Least Squares (PLS) and multiscale approaches, have been reported to be very useful in the task of fault diagnosis of malfunctioning sensors for several types of faults. In this work, we compare the performance of PCA and Multiscale-PCA on a fault based on a change of sensor sensitivity. This type of fault affects chemical gas sensors and it is one of the effects of the sensor poisoning. These two methods will be applied on a dataset composed by the signals of 17 conductive polymer gas sensors, measuring three analytes at several concentration levels during 10 months. Therefore, additionally to performance's comparison, both method's stability along the time will be tested. The comparison between both techniques will be made regarding three aspects; detection, identification of the faulty sensors and correction of faulty sensors response. © 2010 IEEE.
U2 - 10.1109/IJCNN.2010.5596638
DO - 10.1109/IJCNN.2010.5596638
M3 - Book
SN - 9781424469178
T3 - 2010 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS IJCNN 2010IEEE International Joint Conference on Neural Networks (IJCNN)
BT - Fault detection, identification, and reconstruction of faulty chemical gas sensors under drift conditions, using Principal Component Analysis and Multiscale-PCA
PB - IEEE
ER -