A Physics-informed Neural Network for Wind Turbine Main Bearing Fatigue



Yigit A. Yucesan Felipe A. C. Viana


Unexpected main bearing failure on a wind turbine causes unwanted maintenance and increased operation costs (mainly due to crane, parts, labor, and production loss). Unfortunately, historical data indicates that failure can happen far earlier than the component design lives. Root cause analysis investigations have pointed to problems inherent from manufacturing as the major contributor, as well as issues related to event loads (e.g., startups, shutdowns, and emergency stops), extreme environmental conditions, and maintenance practices, among others. Altogether, the multiple failure modes and contributors make modeling the remaining useful life of main bearings a very daunting task. In this paper, we present a
novel physics-informed neural network modeling approach for main bearing fatigue. The proposed approach is fully hybrid and designed to merge physics-informed and data-driven layers within deep neural networks. The result is a cumulative damage model where the physics-informed layers are used model the relatively well-understood physics (L10 fatigue life) and the data-driven layers account for the hard to model components (i.e., grease degradation).

Abstract 30 | PDF Downloads 20



Fatigue Prognosis, Wind Turbine, Physics-informed Machine Learning

Butler, S., O’Connor, F., Farren, D., & Ringwood, J. V. (2012). A feasibility study into prognostics for the main bearing of a wind turbine. In 2012 IEEE International Conference on Control Applications (pp. 1092--1097). Dubrovnik, Croatia: IEEE. doi: 10.1109/ CCA.2012.6402684
Cambron, P., Tahan, A., Masson, C., & Pelletier, F. (2017). Bearing temperature monitoring of a wind turbine using physics-based model. Journal of Quality in Maintenance Engineering, 23(4), 479--488.
Chauhan, S., & Vig, L. (2015, Oct). Anomaly detection in ECG time signals via deep long short-term memory networks. In IEEE International Conference on Data Science and Advanced Analytics (DSAA) (p. 1-7). doi: 10.1109/DSAA.2015.7344872
Chen, T. Q., Rubanova, Y., Bettencourt, J., & Duvenaud, D. K. (2018). Neural ordinary differential equations. In S. Bengio, H. Wallach, H. Larochelle, K. Grauman, N. Cesa-Bianchi, & R. Garnett (Eds.), 31st Advances in Neural Information Processing Systems (pp. 6572--6583). Curran Associates, Inc.
Chung, J., Gulcehre, C., Cho, K., & Bengio, Y. (2015). Gated feedback recurrent neural networks. In International Conference on Machine Learning (pp. 2067--2075).
Connor, J. T., Martin, R. D., & Atlas, L. E. (1994). Recurrent neural networks and robust time series prediction. IEEE Transactions on Neural Networks, 5(2), 240--254. doi: 10.1109/72.279188
Draxl, C., Clifton, A., Hodge, B.-M., & McCaa, J. (2015). The wind integration national dataset (wind) toolkit. Applied Energy, 151(1), 355--366.
Eker, O. F., Camci, F., & Jennions, I. K. (2019). A new hybrid prognostic methodology. International Journal of Prognostics and Health Management, 10(1).
Fatemi, A., & Yang, L. (1998). Cumulative fatigue damage and life prediction theories: a survey of the state of the art for homogeneous materials. International Journal of Fatigue, 20(1), 9 - 34. doi: 10.1016/S0142-1123(97) 00081-9
Frangopol, D. M., Kallen, M.-J., & van Noortwijk, J. M. (2004). Probabilistic models for life-cycle performance of deteriorating structures: review and future directions. Progress in Structural Engineering and Materials, 6(4), 197--212. doi: 10.1002/pse.180 GE-contributors. (2009). GE Energy 1.5 MW Wind Turbine Brochure. Online (retrieved 23 May 2018). Retrieved from https://geosci.uchicago.edu/˜moyer/GEOS24705/Readings/GEA14954C15-MW-Broch.pdf
Goebel, K., Saha, B., & Saxena, A. (2008). A comparison of three data-driven techniques for prognostics. In 62nd Meeting of the Society for Machinery Failure Prevention Technology (pp. 119--131). Virginia Beach, USA: MFPT.
Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning. MIT Press. Retrieved from http://www.deeplearningbook.org
Graves, A., Mohamed, A., & Hinton, G. (2013). Speech recognition with deep recurrent neural networks. In IEEE International Conference on Acoustics, Speech and Signal Processing (pp. 6645--6649). doi: 10.1109/ICASSP.2013.6638947
Gugulothu, N., TV, V., Malhotra, P., Vig, L., Agarwal, P., & Shroff, G. (2018). Predicting remaining useful life using time series embeddings based on recurrent neural networks. International Journal of Prognostics and Health Management, 9(1).
Hill, K. O., & Meltz, G. (1997). Fiber bragg grating technology fundamentals and overview. Journal of Lightwave Technology, 15(8), 1263--1276.
Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8), 1735--1780. doi: 10.1162/neco.1997.9.8.1735
Hornemann, M., & Crowther, A. (2013). Establishing failure modes for bearings in wind turbines. Online (retrieved 23 May 2018). Retrieved from https://www.windpowerengineering.com/mechanical/bearings/establishing-failure-modes-for-bearingsin-wind-turbines/
Iyer, N. S., Qiu, H., Yan, W., & Loparo, K. A. (2007, 05). Early detection of lubrication anomalies in oillubricated bearings. In ASME Turbo Expo 2007: Power for Land, Sea, and Air (pp. 785--794). Montreal, Canada: American Society of Mechanical Engineers Digital Collection. doi: 10.1115/GT2007-27950
Klueber-contributors. (2011). The element that rolls the bearing. Online (retrieved 2 December 2018). Retrieved from https://www.klueber.com/ecomaXL/files/Competence brochurerolling bearing%5B1%5D.pdf
Kohavi, R. (1995). A study of cross-validation and bootstrap for accuracy estimation and model selection. In Proceedings of the 14th International Joint Conference on Artificial Intelligence (pp. 1137--1143). Morgan Kaufmann Publishers Inc.
Liu, Y., & Mahadevan, S. (2009). Efficient methods for timedependent fatigue reliability analysis. AIAA Journal, 47(3), 494--504. doi: 10.2514/1.34331
Lugt, P. M. (2009). A review on grease lubrication in rolling bearings. Tribology Transactions, 52(4), 470--480.
Matei, I., de Kleer, J., Feldman, A., Zhenirovskyy, M., & Rai, R. (2019). Classification based diagnosis: Integrating partial knowledge of the physical system. In 2019 Annual Conference of the PHM Society. Scottsdale, USA: PHM Society. doi: 10.36001/phmconf.2019.v11i1.840
Nascimento, R. G., & Viana, F. A. C. (2019, September). Fleet prognosis with physics-informed recurrent neural networks. In 12th International Workshop on Structural Health Monitoring (pp. 1740--1747). Stanford, USA. doi: 10.12783/shm2019/32301
Raissi, M., & Karniadakis, G. E. (2018). Hidden physics models: machine learning of nonlinear partial differential equations. Journal of Computational Physics, 357, 125 - 141. doi: 10.1016/j.jcp.2017.11.039
Raissi, M., Perdikaris, P., & Karniadakis, G. E. (2018). Numerical Gaussian processes for time-dependent and nonlinear partial differential equations. SIAM Journal on Scientific Computing, 40(1), A172--A198. doi: 10.1137/17M1120762
Roach, D. (2009). Real time crack detection using mountable comparative vacuum monitoring sensors. Smart Structures and Systems, 5(4), 317--328.
Sak, H., Senior, A., & Beaufays, F. (2014). Long shortterm memory recurrent neural network architectures for large scale acoustic modeling. In Fifteenth Annual Conference of the International Speech Communication Association.
Schaeffler-contributors. (2016). Lubrication of rotor bearings in wind turbines. Online (retrieved 2 December 2018). Retrieved from http://www.midpointbearing.com/wp-content/uploads/2017/10/FAG-Main-Shaft-Bearing-Lubrication.pdf
Schober, M., Duvenaud, D. K., & Hennig, P. (2014). Probabilistic ODE solvers with Runge-Kutta means. In Z. Ghahramani, M. Welling, C. Cortes, N. D. Lawrence, & K. Q. Weinberger (Eds.), 27th Advances in Neural Information Processing Systems (pp. 739--747). Curran Associates, Inc.
Sethuraman, L., Guo, Y., & Sheng, S. (2015). Main bearing dynamics in three-point suspension drivetrains for wind turbines. In American wind energy association conference & exhibition. Orlando, USA: AWEA. SKF-contributors. (2007).
SKF spherical roller bearings catalogue. Online (retrieved 5 June 2018). Retrieved from http://www.skf.com/binary/30-148465/6100 EN.pdf
Sutskever, I., Martens, J., & Hinton, G. (2011). Generating text with recurrent neural networks. In 28th International Conference on Machine Learning. Bellevue, USA.
Viana, F. A. C., Nascimento, R. G., Yucesan, Y., & Dourado, A. (2019, Aug). Physics-informed neural networks package. Zenodo. Retrieved from
https://github.com/PML-UCF/pinn doi: 10.5281/zenodo.3356877
Walker, C., & Coble, J. (2018). Wind turbine bearing fault detection using adaptive resampling and order tracking. International Journal of Prognostics and Health Management, 9(2).
Watanabe, F., & Uchida, T. (2015). Micro-siting of wind turbine in complex terrain: simplified fatigue life prediction of main bearing in direct drive wind turbines. Wind Engineering, 39(4), 349--368.
Yu, Y., Yao, H., & Liu, Y. (2018). Physics-based learning for aircraft dynamics simulation. In 2018 Annual Conference of the PHM Society. Philadelphia, USA: PHM Society. doi: 10.36001/phmconf.2018.v10i1.513
Yucesan, Y. (2019). Wind turbine main bearing fatigue life prediction with PINN. https://doi.org/10.7910/DVN/ENNXLZ. Harvard Dataverse. doi: https://doi.org/10.7910/DVN/ENNXLZ
Yucesan, Y., & Viana, F. A. C. (2019a). Onshore wind turbine main bearing reliability and its implications in fleet management. In AIAA SciTech Forum (p. AIAA 2019-1225). San Diego, USA: AIAA. doi: 10.2514/ 6.2019-1225
Yucesan, Y., & Viana, F. A. C. (2019b, Aug). Python scripts for wind turbine main bearing fatigue life estimation with physics-informed neural networks. Zenodo. Retrieved from https://github.com/PML-UCF/pinn wind bearing doi: 10.5281/zenodo.3355725
Zhu, J., Yoon, J. M., He, D., Qu, Y., & Bechhoefer, E. (2013). Lubrication oil condition monitoring and remaining useful life prediction with particle filtering. International Journal of Prognostics and Health Management, 4, 124--138.
Technical Papers