High market competition for sales requires companies to reduce the cost of production if they are to maintain their market shares. Since the cost of maintenance contributes a substantial portion of the production cost, companies must budget maintenance effectively. Machine deterioration prognosis can decrease the cost of maintenance by minimizing the loss of production due to machine breakdown and avoiding the overstocking of spare parts. A new prognostic method is described in this paper which has been developed to forecast the rate of machine deterioration using recurrent neural networks. From tests applying the method to the prediction of nonlinear sunspot activities and vibration based fault trends of several industrial machines, the results have shown that the method is promising. It not only evaluates the seriousness of damage caused by faults, but also forecasts the remaining life span of defective components.

Issue Section:
Research Papers
Topics:
Artificial neural networks, Machinery, Vibration, Maintenance, Damage, Sales, Sunspots
1.
Alguindigue
I. E.
,
Loskiewicz-Buczak
A.
and
Uhrig
R. E.
,
1993
, “
Monitoring and Diagnosis of Rolling Element Bearings Using Artificial Neural Networks
,”
IEEE Transactions on Industrial Electronics
, Vol.
40
, No.
2
, pp.
209
–
217
.
2.
Brockwell, P., and Davis, R. A., 1994, ITSM for Windows: A User’s Guide to Time Series Modeling and Forecasting, Springer Series in Statistics, Springer-Verlag Inc., NY.
3.
Bruel & Kjaer, “Machine Condition Monitoring Using Vibration Analysis,” BA 7095-14, Bruel & Kjaer, Denmark.
4.
Burg, J. P., 1967, “Maximum Entropy Spectral Analysis,” 37th Annual International S.E.G. Meeting, Oklahoma City, Oklahoma.
5.
Cheng, K.L., & Tung, O.W., 1995, “Continuing a Long Line of Monitoring Tradition and Success,” Profile, An International Newsletter From the Monitoring Specialists, Bruel & Kjaer, 3(3), pp. 8–12.
6.
Connor, J.T., Martin, R.D., and Atlas, L.E., 1994, “Recurrent Neural Networks and Robust Time Series Prediction,” IEEE Transactions on Neural Networks, 5(2), pp. 240–254.
7.
Downham, E., & Woods, R., 1971, “It is Dangerous to Associate a Single Vibration Level with Machine Performance,” Proceedings of the ASME Conference, Vibration 96, No. 71, Toronto.
8.
Faller
W. E.
, and
Schreck
S. J.
,
1995
, “
Real-Time Prediction of Unsteady Aerodynamics: Application for Aircraft Control and Manoeuvrability Enhancement
,”
IEEE Transactions on Neural Networks
, Vol.
6
, pp.
1461
–
1468
.
9.
Gabr, M. M., and Rao, T. S., 1981, “The Estimation and Prediction of Subset Bilinear Time Series Models with Applications,” Journal of Time Series Analysis, 2 (155).
10.
Groot, C., and Wu¨rtz, D., 1991, “Analysis of Univariate Time Series with Connectionist Nets: A Case Study of Two Classical Examples,” Neurocomputing 3, pp. 177–192.
11.
Fausett, L., 1994, Fundamentals of Neural Networks, Prentice Hall.
12.
Izenman A.J. and Wolf J.R., 1985, “The Zurich Sunspot Relative Numbers,” Math. Intelligencer, 7.
13.
Kuo, Jyh-Ming, Principe, J.C, and de Vries, B., 1992, “Prediction of Chaotic Time Series Using Recurrent Neural Networks,” Neural Networks for Signal Processing II: Proceedings of the 1992 IEEE-SP, New York, pp. 436–443.
14.
Press, W. H., Flannery, B.P., and Teukolsky, S.A., 1986, Numerical Receipts, Cambridge University Press, Cambridge.
15.
Sakamoto, Y., Ishiguro, M., and Kitagawa, G., 1986, Akaike Information Criterion Statistics, KTK Scientific Publishing, Japan.
16.
Saxen, B., 1995, NNDT- A Neural Network Development Tool: User’s Guide, Abo Akademi University, Finland.
17.
Schalkoff, R., 1997, Artificial Neural Networks, McGraw-Hill Inc., NY, pp. 246–307.
18.
Sorsa, T., Koivo H. N., and Koivisto, H., 1991 “Neural Networks in Process Fault Diagnosis,” IEEE Trans, on System, Man, and Cybemetica, Vol. 21, No. 4.
19.
Strudwick, S.D., & Perryman, R., 1993, “Trend Analysis of Remote Standby Generating Sets Using Neural Networks,” Technique and Application of Neural Networks, Taylor M. & Lisboa P., ed. Ellis Horwood, New York, pp. 118–132.
20.
Tong, H. and Lim, S.K., 1980, “Threshold Autoregression, Limited Cycles and Cyclical Data,”. Journal of R. Stat. Soc. B42, (245).
21.
Trippi, R., & Turban, E., 1996, Neural Networks in Finance and Investing, Irwin Professional Publishing, Chicago, pp. 724–731.
22.
Tse
P.
,
Atherton
D.
, and
Wang
D.
,
1996
, “
Harmony Theory Yields Robust Machine Fault Diagnostic System Based on Learning Vector Quantization Classifier
,”
Engineering Applications of Artificial Intelligence, Elsevier Science Inc.
, Vol.
9
, No.
5
, pp.
487
–
498
.
23.
Tse P., and Wang, D., 1996 “A Hybrid Neural Networks based Machine Condition Forecaster and Classifier by using Multiple Variable Parameters,” Proceedings of the IEEE ICNN’96, Washington, DC, USA, Vol. 4/4, pp. 2096–2100.
24.
Weigend, W., &. Gershenfeld, N., 1993, “Results of the Time Series Prediction Competition at the Sante Fe Institute,” Proceedings of the IEEE ICNN’93, San Francisco, USA, pp. 1786–1793.
25.
Ziemke, T., Boden M., and Niklasson, L., 1997, “Oil Spill Detection: a Case Study Using Recurrent Neural Networks,” Browne A., ed. Neural Networks Analysis, Architectures and Applications, Institute of Physics Publishing, Ltd., Philadalphia, pp. 224–250.
This content is only available via PDF.
Copyright © 1999
 by The American Society of Mechanical Engineers
You do not currently have access to this content.