Wind turbine manufacturers have adopted condition monitoring systems to monitor and report a turbine’s health and operating parameters to ensure that the system operates within its design specifications. While the present systems use specialized condition monitoring hardware to detect
abnormal acoustic or vibration signals, it is not capable of pinpointing the exact location of the fault apart from isolating the system from which the signal originated. This drawback can be attributed to the requirement of powerful signal processors in order to decode the signal and efforts to train a system to identify the signal emitted by a faulty component. In the light of recent advancement of datadriven approaches and signal processing, these drawbacks can be overcome with increased computation power and sophisticated algorithms that foray into every integrated
system. This paper reports such an investigation conducted on a miniature wind turbine planetary gearbox subjected to multi-component failures. The vibration signals were acquired using two accelerometers placed inside the gearbox. The speed of the gearbox was varied according to a
simulated wind flow pattern. The primary goal of the study was to investigate the practicality of implementing datadriven approaches to categorise multi-component faults from a composite non-stationary signal. Short time Fourier transforms (STFT) coefficients were used as attributes by a
set of data-driven algorithms to build machine learning models. Each model built was tested with a randomised set of instances which was reserved from the main dataset and tested multiple times by means of cross validation. The novelty in the paper entails a methodology which has been devised to classify faults using a randomised vibration dataset with little human intervention by means of machine learning algorithms. The authors propose that this methodology can also be used for real-time fault detection and classification for various machinery and components.
condition monitoring, data driven methods, Wind Turbine, Automated fault detection
Elangovan, M., Sugumaran, V., Ramachandran, K. I., & Ravikumar, S. (2011). Effect of SVM kernel functions on classification of vibration signals of a single point cutting tool. Expert Systems with Applications, 38(12), 15202–15207. https://doi.org/10.1016/j.eswa.2011.05. 081
Evans, M.-H. (2012). White structure flaking (WSF) in wind turbine gearbox bearings: effects of ‘butterflies’ and white etching cracks (WECs). Materials Science and Technology, 28(1), 3–22. https://doi.org/10.1179/026708311X13135950699254
Feng, Z., Lin, X., & Zuo, M. J. (2016). Joint amplitude and frequency demodulation analysis based on intrinsic time-scale decomposition for planetary gearbox fault diagnosis. Mechanical Systems and Signal Processing, 72–73, 223–240. https://doi.org/10.1016/j.ymssp. 2015.11.024
Frank, E., Hall, M. A., & Witten, I. H. (2016). The WEKA Workbench. Morgan Kaufmann, Fourth Edition, 553–571. https://doi.org10.1016B978-0-12-804291-5.00024-6
Gryllias, K., Andre, H., Leclere, Q., & Antoni, J. (2017). Condition monitoring of rotating machinery under varying operating conditions based on Cyclo-Non-Stationary Indicators and a multi-order probabilistic approach for Instantaneous Angular Speed tracking. IFAC-PapersOnLine, 50(1), 4708–4713. https://doi.org/10.1016/j.ifacol.2017.08.857
Haastrup, M., Hansen, M. R., & Ebbesen, M. K. (2011). Modeling of wind turbine gearbox mounting. Modeling, Identification and Control, 32(4), 141–149. https://doi.org/10.4173/mic.2011.4.2
Lei, Y., & Zuo, M. J. (2009). Gear crack level identification based on weighted K nearest neighbor classification algorithm. Mechanical Systems and Signal Processing, 23(5), 1535–1547. https://doi.org/10.1016/j.ymssp. 2009.01.009
Lu, B., Li, Y., Wu, X., & Yang, Z. (2009). A review of recent advances in wind turbine condition monitoring and fault diagnosis. Electronics and Machines in Wind, 1–7. https://doi.org/10.1109/PEMWA.2009.5208325
Moosavian, A., Najafi, G., Ghobadian, B., & Mirsalim, M. (2017). The effect of piston scratching fault on the vibration behavior of an IC engine. Applied Acoustics, 126, 91–100. https://doi.org/10.1016/j.apacoust. 2017.05.017
Muralidharan, V., & Sugumaran, V. (2013). Feature extraction using wavelets and classification through decision tree algorithm for fault diagnosis of monoblock centrifugal pump. Measurement, 46(1), 353–359.https://doi.org/10.1016/j.measurement.2012.07.007
Ozer, I., Ozer, Z., & Findik, O. (2017). Lanczos kernel based spectrogram image features for sound classification. Procedia Computer Science, 111(2015), 137–144. https://doi.org/10.1016/j.procs.2017.06.020
Rajeswari, C., Sathiyabhama, B., Devendiran, S., & Manivannan, K. (2014). A Gear fault identification using wavelet transform, rough set based GA, ANN and C4.5 algorithm. Procedia Engineering, 97, 1831–1841. https://doi.org/10.1016/j.proeng.2014.12.337.
Quinlan, J. R(1996). Improved Use of Continuous Attributes in C4 . 5, Journal of Artificial Intelligence Research, 4, 77-90. https://doi.org/10.1613/jair.279
Sheng, S. (2015). Wind Turbine Gearbox Reliability Database, Condition Monitoring, and O&M Research Update. PR-5000-63868, National Renewable Energy Laboratory (NREL), Golden, CO (US),.
Shuangwen Sheng and Paul Veers. (2011). Wind Turbine Drivetrain Condition Monitoring - An Overview. Machinery Failure Prevention Technology (MFPT): The Applied Systems Health Management Conference 2011, (October).
Singh, A., & Parey, A. (2017). Gearbox fault diagnosis under non-stationary conditions with independent angular re-sampling technique applied to vibration and sound emission signals. Applied Acoustics. https://doi.org/10.1016/j.apacoust.2017.04.015
Teng, W., Ding, X., Zhang, X., Liu, Y., & Ma, Z. (2016). Multi-fault detection and failure analysis of wind turbine gearbox using complex wavelet transform. Renewable Energy, 93, 591–598. https://doi.org/10.1016/j.renene.2016.03.025
Villa, L. F., Reñones, A., Perán, J. R., & De Miguel, L. J. (2012). Statistical fault diagnosis based on vibration analysis for gear test-bench under non-stationary conditions of speed and load. Mechanical Systems and Signal Processing, , 29, 436–446. https://doi.org/10.1016/j.ymssp.2011.12.013
Wang, Z., Wang, J., & Wang, Y. (2018). An intelligent diagnosis scheme based on generative adversarial learning deep neural networks and its application to planetary gearbox fault pattern recognition. Neurocomputing. https://doi.org/10.1016/j.neucom. 2018.05.024
Xie, Z., Mcloughlin, I., Zhang, H., Song, Y., & Xiao, W. (2016). A new variance-based approach for discriminative feature extraction in machine hearing
classification using spectrogram features. Digital Signal Processing: A Review Journal, 54, 119–128. https://doi.org/10.1016/j.dsp.2016.04.005
Yang, W. (2013). Condition Monitoring the Drive Train of a Direct Drive Permanent Magnet Wind Turbine Using Generator Electrical Signals. Journal of Solar Energy Engineering, 136(2), 021008. https://doi.org/10.1115/1.4024983
Zhang, W., Han, J., & Deng, S. (2017). Heart sound classification based on scaled spectrogram and tensor decomposition. Expert Systems with Applications, 84, 220–231. https://doi.org/10.1016/j.eswa.2017.05.014
Zimroz, R., Bartelmus, W., Barszcz, T., & Urbanek, J. (2014). Diagnostics of bearings in presence of strong operating conditions non-stationarity - A procedure of load-dependent features processing with application to wind turbine bearings. Mechanical Systems and Signal Processing, 46(1), 16–27. https://doi.org/10.1016/j.ymssp.2013.09.010