This chapter presents an improved incipient FD method based on Kullback-Leibler divergence (KLD) under multivariate statistical analysis frame. Different from the traditional MVA-based FD methods, this methodology can detect slight anomalous behaviors by comparing the PDF online with the reference PDF obtained from large scale off-line datasets.
PCA and Kullback-Leibler Divergence-Based FDD Methods
Lecture Notes in Intelligent Transportation and Infrastructure
Data-driven Detection and Diagnosis of Faults in Traction Systems of High-speed Trains ; Chapter : 7 ; 119-135
2020-04-26
17 pages
Article/Chapter (Book)
Electronic Resource
English
Ellipticity and Circularity Measuring via Kullback–Leibler Divergence
British Library Online Contents | 2016
|IEEE | 2003
|Minimization of the Kullback-Leibler Divergence for Nonlinear Estimation
Online Contents | 2017
|British Library Conference Proceedings | 2003
|