Automatic Threshold Setting and Its Uncertainty Quantification in Wind Turbine Condition Monitoring System

Setting optimal alarm thresholds in vibration based condition monitoring system is inherently difficult. There are no established thresholds for many vibration based measurements. Most of the time, the thresholds are set based on statistics of the collected data available. Often times the underlying probability distribution that describes the data is not known. Choosing an incorrect distribution to describe the data and then setting up thresholds based on the chosen distribution could result in sub-optimal thresholds. Moreover, in wind turbine applications the collected data available may not represent the whole operating conditions of a turbine, which results in uncertainty in the parameters of the fitted probability distribution and the thresholds calculated. In this study, Johnson, Normal, and Weibull distributions are investigated; which distribution can best fit vibration data collected from a period of time. False alarm rate resulted from using threshold determined from each distribution is used as a measure to determine which distribution is the most appropriate. This study shows that using Johnson distribution can eliminate testing or fitting various distributions to the data, and have more direct approach to obtain optimal thresholds. To quantify uncertainty in the thresholds due to limited data, implementations with bootstrap method and Bayesian inference are investigated.

[1]  C. J. Crabtree.  Condition monitoring techniques for wind turbines , 2011 .

[2]  E. Bechhoefer,et al.  Use of non-Gaussian distribution for analysis of shaft components , 2006, 2006 IEEE Aerospace Conference.

[3]  Francis A. Roesch,et al.  Bayesian estimation for the three-parameter Weibull distribution with tree diameter data , 1994 .

[4]  E. Bechhoefer,et al.  A Generalized Process for Optimal Threshold Setting in HUMS , 2007, 2007 IEEE Aerospace Conference.

[5]  David He,et al.  Gear Health Threshold Setting Based On a Probability of False Alarm , 2011 .

[6]  Peter Jamieson,et al.  Offshore Wind Turbines , 2011 .

[7]  Roman Wisznia.  Condition Monitoring of Offshore Wind Turbines , 2013 .

[8]  G. C. Tiao,et al.  Bayesian inference in statistical analysis , 1973 .

[9]  Tomasz Barszcz,et al.  Modeling of probability distribution functions for automatic threshold calculation in condition monitoring systems , 2013 .

[10]  David J. C. MacKay,et al.  Information Theory, Inference, and Learning Algorithms , 2004, IEEE Transactions on Information Theory.

[11]  J. R. Wilson,et al.  Modeling input processes with Johnson distributions , 1989, WSC '89.

[12]  C. Cempel,et al.  Limit value in the practice of machine vibration diagnostics , 1990 .

[13]  N. L. Johnson,et al.  Systems of frequency curves generated by methods of translation. , 1949, Biometrika.

[14]  Kun S. Marhadi,et al.  Quantifying uncertainty in statistical distribution of small sample data using Bayesian inference of unbounded Johnson distribution , 2008 .

[15]  Czesław Cempel,et al.  Simple condition forecasting techniques in vibroacoustical diagnostics , 1987 .

[16]  E. Bechhoefer,et al.  Setting HUMS condition indicator thresholds by modeling aircraft and torque band variance , 2004, 2004 IEEE Aerospace Conference Proceedings (IEEE Cat. No.04TH8720).