Hybrid HMM/SVM method for predicting cutting chatter

Yongtao Jiang; Chunliang Zhang

doi:10.1117/12.716150

13 October 2006 Hybrid HMM/SVM method for predicting cutting chatter

Yongtao Jiang, Chunliang Zhang

Proceedings Volume 6280, Third International Symposium on Precision Mechanical Measurements; 62801Q (2006) https://doi.org/10.1117/12.716150
Event: Third International Symposium on Precision Mechanical Measurements, 2006, Urumqi, China

Abstract

According to the properties of cutting chatter, a new chatter forecast system has been developed based on Hidden Markov Model (HMM) and Support Vector Machine (SVM). This system uses HMM as the recognition method and SVM as the prediction method. At the same time, means like wavelet package decomposition are also employed to extract the cutting features. The basic idea and general steps of this method are as follow. Firstly, the cutting signals are analyzed step by step in the same interval using wavelet packet decomposition. Secondly, the energy in every spectrum section are calculated and scaled in order to get general property. As a result, the energy distribution information and energy transition curve of different spectrum section can be retrieved. Then, SVR algorithm is applied to predict the trend of energy transition. The results, after scalar quantized, at last are input into HMMs to determine whether in chatter period. Certainly, current state still needs to be distinguished. The simulation results indicate that the new predicting method has good discriminating performances and high forecast accuracy.

Citation Download Citation

Yongtao Jiang and Chunliang Zhang "Hybrid HMM/SVM method for predicting cutting chatter", Proc. SPIE 6280, Third International Symposium on Precision Mechanical Measurements, 62801Q (13 October 2006); https://doi.org/10.1117/12.716150

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available