scholarly journals Compression-based methods for nonparametric density estimation, on-line prediction, regression and classification for time series

Author(s):  
Boris Ryabko
2013 ◽  
Vol 819 ◽  
pp. 160-164
Author(s):  
Yong Xiang Jiang ◽  
Bing Du ◽  
Pan Zhang ◽  
San Peng Deng ◽  
Yu Ming Qi

On-line monitoring recognition for machining chatter is one of the key technologies in manufacturing. Based on the nonlinear chaotic control theory, the vibration signal discrete time series for on-line monitoring indicator is studed. As in chatter the chaotic dynamics process attractor dimension is reduced, the KolmogorovSinai entropy (K-S) index is extracted to reflected the regularity of workpiece chatter, then the k-S entropy is simplified by coarse - grained entropy rate (CER), which can easily evaluated as chatter online monitoring threshold value. The milling test shows that the CER have a sharp decline when chatter occurre, and can quickly and accurately forecast chatter.


2019 ◽  
Vol 88 ◽  
pp. 506-517 ◽  
Author(s):  
Izaskun Oregi ◽  
Aritz Pérez ◽  
Javier Del Ser ◽  
Jose A. Lozano

Sign in / Sign up

Export Citation Format

Share Document