Abstract
Kernel entropy component analysis (KECA) is a new method to extract features and perform dimension reduction by preserving most of Renyi entropy. However,it appears to have deficiencies. The entropy estimation in KECA may be not the most accurate and it is sensitive to a single kernel radius. This paper proposes a multiplicative bias correction method of the entropy estimation. According to the analysis of the relation of statistics in the kernel feature space,we introduces two kernel radii to make KECA less sensitive to the kernel radius.A method for fault diagnosis of analog circuits based on the combination of improved KECA and extreme learning machine (ELM) is presented.Through wavelet decomposition of sampled signals,features are extracted, following improved KECA for feature dimension reduction.Next the fault patterns are classified by ELM.Case studies on two analog circuits demonstrating our diagnostics method are presented.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
