×

A comparative study of three artificial neural networks for the detection and classification of gear faults. (English) Zbl 1078.68681

Summary: Artificial neural networks (ANN) have been recognized as a powerful tool for classification and pattern recognition in various fields of applications. This paper presents an overview of three ANN architectures and the results of applying those ANNs for the detection and classification of malfunction, wear and damage of a gearbox operating under steady state conditions. The ANN models studied are: feed forward back propagation (FFBP), functional link network (FLN) and learning vector quantization (LVQ). Three artificial defects were deliberately introduced to the gearbox and these are: (1) loose key, (2) single tooth flank wear and (3) full tooth breakage (missing tooth). Vibration signals, collected from extensive experimentation, were analyzed using time and frequency domain descriptors that were used as feature vectors to feed the ANNs. The results show that, for this study, the FLN learns more quickly and is more accurate in operation than the FFBP or the LVQ. The LVQ algorithm exhibits faster rate of convergence than the FFBP but suffers more from misclassifications.

MSC:

68T05 Learning and adaptive systems in artificial intelligence
68T10 Pattern recognition, speech recognition
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Abu-Mahfouz IA, Proc. of the SAE Noise and Vibration Conference and Exposition, Traverse City, Michigan (2001)
[2] Bart K, Neural Networks and Fuzzy Systems (1992)
[3] Corner JT, IEEE Trans. Neural Networks 5 pp 240– (1994) · doi:10.1109/72.279188
[4] Faller WE, IEEE Trans. Neural Networks 6 pp 1461– (1995) · doi:10.1109/72.471362
[5] Forrester BD, Time–Frequency Signal Analysis pp 406– (1992)
[6] Freeman JA, Neural Networks: Algorithms, Applications, and Programming Techniques (1991)
[7] James CL, Wear 241 pp 26– (2000) · doi:10.1016/S0043-1648(00)00356-2
[8] Kohonen T, Technical Report TKK-F-A601, Helsinki University of Technology (1986)
[9] Lin ST, Mech. Syst. Signal Process. 11 pp 603– (1997) · doi:10.1006/mssp.1997.0097
[10] Looney CG, Pattern Recognition Using Neural Networks, Theory and Algorithms for Engineers and Scientists (1997)
[11] Signal Processing Toolbox, User’s Guide (1998)
[12] Naim B, Mech. Syst. Signal Process. 15 pp 1091– (2001) · doi:10.1006/mssp.2000.1338
[13] Pao YH, Adaptive Pattern Recognition and Neural Network (1989)
[14] Schalkoff R, Artificial Neural Networks (1997)
[15] Trippi R, Neural Networks in Finance and Investing (1996)
[16] Tse PW, ASME Trans. J. Vibration Acoustics 121 pp 355– (1999) · doi:10.1115/1.2893988
[17] Wang WJ, Mech. Syst. Signal Process. 7 pp 193– (1993) · doi:10.1006/mssp.1993.1008
[18] DOI: 10.1006/jsvi.1996.0226 · doi:10.1006/jsvi.1996.0226
[19] Williams WJ, Mech. Syst. Signal Process. 14 pp 545– (2000) · doi:10.1006/mssp.2000.1296
[20] Wilson QW, Mech. Syst. Signal Process. 15 pp 905– (2001) · doi:10.1006/mssp.2001.1392
[21] Ziemke T, Neural Networks Analysis, Architectures and Applications pp 224– (1997)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.