Diagnostic and Prognostic Modeling of High-Speed Milling Machine Cutters
This paper investigates diagnostic and prognostic modeling for high-speed end mill cutters. Some key issues related to the development of such models are addressed, including degradation feature extraction and statistical inference for diagnostics and prognostics. In particular, Auto-associative Kernel Regression, as a non-parametric approach, is used to detect tool wear, and a Weibull proportional hazard model enhanced by a Bayesian updating method is proposed for prognostics. Experimental results indicate that these methods along with carefully selected degradation features provide effective diagnostic and prognostic tools for such applications.
Machine tool condition monitoring Empirical model Prognostic model
SEYED AHMAD NIKNAM HAITAO LIAO
Department of Industrial & Information Engineering, The University of Tennessee, Knoxville, Tennessee 37996, USA
国际会议
北京
英文
201-206
2011-06-20（万方平台首次上网日期，不代表论文的发表时间）