Kernelized relative entropy for direct fault detection in industrial rotary kilns
- 10 April 2018
- journal article
- research article
- Published by Wiley in International Journal of Adaptive Control and Signal Processing
- Vol. 32 (7) , 967-979
Abstract
Summary: The objective of this work is to use a 1‐dimensional signal that reflects the dissimilarity between multidimensional probability densities for detection. With the modified Kullback‐Leibler divergence, faults can be directly detected without any normality assumption or joint monitoring of related test statistics in different subspaces such as theT2andSPEin principal component analysis–based methods. To relieve the difficulty associated with asymptotic high‐dimensional density estimates, we have estimated the density ratio rather than the densities themselves. This can be done by approximating the density ratio with kernel basis functions and learn the weights from the available data. The developed algorithm is generic and can be applied to any industrial system as long as process historical data is available. As a case study, we apply this algorithm to a real rotary kiln in operation, which is an integral part of the cement manufacturing plant of Ain El Kebira, Algeria.Keywords
This publication has 40 references indexed in Scilit:
- Change-point detection in time-series data by relative density-ratio estimationNeural Networks, 2013
- Distributions of the Kullback–Leibler divergence with applicationsBritish Journal of Mathematical and Statistical Psychology, 2011
- Identification, prediction and detection of the process fault in a cement rotary kiln by locally linear neuro-fuzzy techniqueJournal of Process Control, 2010
- Testing for Homogeneity in Mixture Using Weighted Relative EntropyCommunications in Statistics - Simulation and Computation, 2008
- Tool condition monitoring in milling using vibration analysisInternational Journal of Production Research, 2007
- Test Construction for Cognitive DiagnosisApplied Psychological Measurement, 2005
- Recursive partial least squares algorithms for monitoring complex industrial processesControl Engineering Practice, 2003
- Information and metrics in Hilbert spacePhysical Review A, 1997
- The marginalized likelihood ratio test for detecting abrupt changesIEEE Transactions on Automatic Control, 1996
- On Information and SufficiencyThe Annals of Mathematical Statistics, 1951