Development and application of a statistically based feature extraction algorithm for monitoring tool wear in Circuit Board Assembly

Vibroacoustic signals of rotating machinery are composed of sums of modulated periodicities, broadband random components, and occasionally a set of transient responses. These signals are not ergodic as the modulated periodicities are partially coherent. Progressive wear of the rotating machine causes the nonlinear structure of the received signal to intensify, and nonlinearity results in transfer of energy between harmonics of the signal's periodic components. Statistics developed from bispectrum and second-order cumulant spectrum estimates of the measured signal are combined with power spectrum amplitudes as feature inputs for standard multivariate classifiers. The higher-order statistics measure, respectively, the extent of nonlinearity and intermodulation of the received signal. Classification results of simulated and actual incipient wear data collected from a controlled experiment drilling circuit boards illustrate the potential of this novel statistical signal processing approach.

[1]  N. L. Johnson,et al.  Multivariate Analysis , 1958, Nature.

[2]  X. D. Fang,et al.  Comprehensive Tool Wear Estimation in Finish-Machining via Multivariate Time-Series Analysis of 3-D Cutting Forces , 1990 .

[3]  Richard W. Barker.  Incipient Fault Detection Using Higher-Order Statistics , 1991 .

[4]  Yusuf Altintas,et al.  In-process detection of tool breakages using time series monitoring of cutting forces , 1988 .

[5]  T. I. Liu,et al.  On-Line Detection of Drill Wear , 1990 .

[6]  Y. S. Tarng,et al.  Sensing Cutter Breakage in Milling , 1988 .

[7]  Ciro Neal Ramirez,et al.  Drill Wear Monitoring in Circuit Board Manufacturing: An Experimental Study , 1992 .

[8]  Melvin J. Hinich,et al.  Detecting a transient signal by bispectral analysis , 1990, IEEE Trans. Acoust. Speech Signal Process..

[9]  M.R. Raghuveer,et al.  Bispectrum estimation: A digital signal processing framework , 1987, Proceedings of the IEEE.

[10]  Y. S. Tarng.  Study of milling cutting force pulsation applied to the detection of tool breakage , 1990 .

[11]  J. Lacoume.  Higher order statistics : proceedings of the International Signal Processing Workshop on Higher Order Statistics, Chamrousse, France, July 10-12, 1991 , 1992 .

[12]  M. Hinich.  Higher order cumulants and cumulant spectra , 1990, Fifth ASSP Workshop on Spectrum Estimation and Modeling.

[13]  Ekkard Brinksmeier,et al.  Prediction of Tool Fracture in Drilling , 1990 .

[14]  Ciro Neal Ramirez.  On-line drill condition monitoring by measurement of drill spindle vibration and dynamics , 1991 .

[15]  M. A. Elbestawi,et al.  Process monitoring in milling by pattern recognition , 1989 .

[16]  Yukitatsu Kashimura.  Study on detection of drill wear and breakage (1st report). Development of monitoring system , 1984 .

[17]  M. Hundal.  Mechanical signature analysis , 1983 .

[18]  M. Hinich.  Testing for Gaussianity and Linearity of a Stationary Time Series. , 1982 .

[19]  Jerry M. Mendel,et al.  Workshop on Higher-Order Spectral Analysis Held at Vail, Colorado on 28- 30 June 1989 , 1989 .

[20]  David Dornfeld,et al.  Sensor Integration Using Neural Networks for Intelligent Tool Condition Monitoring , 1990 .