An automated low-cost condition monitoring system for quality control of automotive speedometers

Abstract

The growing demands for high-quality products with improved specifications increase the requirements for condition monitoring and quality evaluation to ensure that no defective or below-specification products reach the customer. This paper presents a low-cost condition-monitoring system using acoustics to evaluate the quality of mechanically driven automotive speedometers during manufacture. Signal conditioning methods employ discrete cosine transformation to simplify the frequency spectrum of the signals. A novelty detection approach is optimized to characterize the ‘normal’ conditions of the speedometers and to reject any speedometers that are not so categorized on the basis of the sound waves produced from experimental running of the speedometers. The suggested techniques are found useful in developing an automated condition-monitoring system for deselecting low-quality automotive speedometers.

Keywords

mechatronics automotive speedometers condition monitoring artificial intelligence quality control

References

Elick

Yoder

Automating speedometer inspections. Robotics World, June 1987, 5(6), 22–25.

Mannan

M. A.

Kassim

Jing

Application of image and sound analysis techniques to monitor the condition of cutting tools. Pattern Recognition Lett., October 2000, 21(11), 969–976.

Heng

R. B. W.

Nor

M. J. M.

Statistical analysis and vibration signals for monitoring rolling element bearing condition. Appl. Acoust., January-March 1998, 53(1–3), 211–226.

Carney

M. S.

Mann

J. A.

Gagliardi

Adaptive filtering of sound pressure signals for monitoring machinery in noisy environments. Appl. Acoust., 1994, 43(4), 333–351.

Belchamber

Collins

Sound processing for your plant?

Control Instrumentation, October 1993, 25(10), 41–42.

Devaraj

Rajnarayanan

S. B.

Senthilnathan

Automating speedometer calibration. Evaluation Engng, 2000, 39(11), 110–131.

Bishop

C. M.

Neural Networks for Pattern Recognition, 1995 (Clarendon, Oxford).

Roberts

S. J.

Extreme value statistics for novelty detection in biomedical data processing. Proc. Instn Electl Engrs: Sci. Measmt Technol., November 2000, 147(6), 363–367.

Albrecht

Bush

Kloppenburg

Metze

Tavan

Generalised radial basis function networks for classification and novelty detection: Self-organisation of optimal Bayesian decision. Neural Networks, December 2000, 13(10), 1075–1093.

10.

Manson

Pierce

S. G.

Worden

Monnier

Guy

Atherton

Long-term stability of normal condition data for novelty detection. Proc. SPIE 2000, 3985, 323–334.

11.

Zorriassatine

Application of neural networks for detection of special causes in multivariate statistical process control. PhD thesis, University of Nottingham, April 2000.

12.

Fukunaga

Introduction to Statistical Pattern Recognition, 2nd edition, 1990 (Academic Press, London).

13.

Parzen

Stochastic Processes, 1962 (Holden-Day, San Francisco, California).

14.

Specht

D. F.

Probabilistic neural networks. Neural Networks, 1990, 3(1), 109–118.

15.

Nabney

Bishop

C. M.

Netlab neural network software. http://www.ncrg.aston.ac.uk/netlab/index.html, Neural Computing Research Group, Information Engineering, Aston University, Birmingham B4 7ET, UK, October 2000.

16.

Haykin

Neural Networks—A Comprehensive Foundation, 1999 (Prentice-Hall, Englewood Cliffs, New Jersey).

17.

Demuth

Beale

Neural Network Toolbox for Use with MATLAB, User's Guide, 1994 (The Math Works Inc., Natick, Massachusetts).

18.

Pao

Y. H.

Adaptive Pattern Recognition and Neural Networks, 1989 (Addison-Wesley, Reading, Massachusetts).

19.

Kohonen

Self-Organization and Associative Memory, 3rd edition, 1989 (Springer, Berlin).

20.

Newland

D. E.

An Introduction to Random Vibrations, Spectral and Wavelet Analysis, 1993 (Longman, London).

21.

Al-Habaibeh

Gindy

A new approach for systematic design of condition monitoring systems for milling processes. J. Mater. Processing Technol., November 2000, 107(1–3), 243–251.

22.

Jain

A. K.

Fundamentals of Digital Image Processing, 1989 (Prentice-Hall, Englewood Cliffs, New Jersey).