Some of the problems involved with monitoring the variable vibration behaviour observed on large turbine generators are described. The uses of neural networks are discussed and an alternative to the more commonly used network is proposed which has a significantly shorter training time. Examples are given of the use of these techniques for studying an oil whirl problem, an intertum fault, and reactive load dependent behaviour.
Get full access to this article
View all access options for this article.
References
1.
KorteniemiA.An expert system for fault diagnostics of rotating plant. ESAP, Seattle, 1989.
2.
FutterD. NHewittG.MayesI. W.TeeT. K.Techniques for remote monitoring and processing of large quantities of on-line vibration data. IMechE Conference on Vibrations in rotating machinery. University of Bath, 7–10 September 1992, paper C432/082, pp. 425–433 (Mechanical Engineering Publications, London).
3.
ZhugeQ.LuY.Signature analysis for reciprocating machinery with adaptive signal processing. Proc. Instn Mech. Engrs, Part C, 1991, 205(C5), 305–310.
4.
GasperB.CorneliousS.HerbertR.Analysis of run down frequency response component data using modal analysis techniques. IMechE Conference on Vibrations in rotating machinery, Heriot-Watt University, Edinburgh, 13–15 September 1988 (Mechanical Engineering Publications, London).
5.
HerbertR. G.Turbine alternator rundown vibration analysis automatic crack detection. ASME Conference on Vibration and noise, Boston, 1987 (American Society of Mechanical Engineers, New York).
6.
SkirtP. J. C.WhitcombR. C.The analysis of the acoustic emission of jet engines using artificial neural networks. Condition Monitoring and Diagnostic Technol., 1990, 1..
7.
KeyvanS.DurgA.Pump shaft condition monitoring via artifical neural networks. Fifth Predictive Maintenance Conference, September 1992 (EPRI, Knoxville).
8.
AlguindigneI. E.Loskiewicz-BuczakA.Vibration monitoring using sensor readings predicted by a neural network. Fifth Predictive Maintenance Conference, September 1992 (EPRI, Knoxville).
9.
IEEE Trans. on Acoustics, Speech, and Signal Processing, 1988, 36(7).
10.
WidrowB.LehrM.30 years of adaptive neural networks: perception, madaline, and back propagation. Proc. IEEE, 1990, 78(9).
11.
BillingsS. A.ZhuQ. M.Rational model identification using an extended least-squares algorithm. Int. J. Control, 1991, 54(3).
12.
FunahashiK.On the approximate realisation of continuous mapping by neural networks. Neural Networks, 1982, 2.
13.
BaldiP.HornikK.Neural networks and principal component analysis. INNS Neural Networks, 1982, 2(1).
14.
HewittG.MayesI. W.Online vibration monitoring of turbine generators using principle component analysis and the generalised T statistic. Internal Report, TR/91/14072/R, 1991.
15.
RumelhartD. E.HintonG. E.WilliamsR. J.Learning internal representations by error propagation. In Parallel distributed processing, 1987, pp. 318–362 (MIT Press, Cambridge, Mass.).
16.
ChenS.CowanC. F. N.BillingsS. A.GrantP. M.Parallel recursive prediction error algorithm for training layered neural networks. Int. J. Control, 1990, 51(6).
17.
PowellM. I. D.Radial basis functions for multivariate interpolation: A review. In IMA Confon algorithms for the approximation of functions and data, 1987 (RMCS, Shrivenham).
18.
MicchelliC. A.Interpolation of scattered data: distance matrices and conditionally positive definite functions. Constructive Approximations, 1986, 2, 11–22.
ChenS.BillingsS. A.GrantP. M.Recursive hybrid algorithm for non-linear system identification using radial basis function networks. Int. J. Control, 1992, 55(5).
21.
DudaR. O.HartP. N. E.Pattern classification and scene analysis, 1973 (John Wiley, Chichester).
22.
AndersonT. W.An introduction to multivariate statistical analysis, 1958 (John Wiley, New York).
23.
ChenS.BillingsS. A.CowanC. F. N.GrantP. M.Practical identification of NARMAX models using radial basis functions. Int. J. Control, 1990, 52(6).
24.
ChenY.Machine condition monitoring by inverse filtering and statistical analysis. Mechanical Systems and Signal Processing, 1992, 6(2), 177–189 (Power Technology Department, PowerGen plc).
