A novel Architecture for on-line Failure Prognosis using Probabilistic Least Squares Support Vector Regression Machines

##plugins.themes.bootstrap3.article.main##

##plugins.themes.bootstrap3.article.sidebar##

Published Mar 26, 2021
Taimoor Khawaja Dr. George Vachtsevanos

Abstract

The ability to forecast machinery failure is vital to reducing maintenance costs, operation downtime and safety hazards. Recent advances in condition monitoring technologies have given rise to a number of prognostic schemes (both model-based and data-driven) that attempt to forecast machinery health by constructing health propagation models for the underlying systems. In particular, algorithms that use the data-driven approach learn models directly from the data, rather than using a hand-built model based on human expertise. This paper introduces a novel architecture for data-driven Failure Prognosis of complex non-linear systems using Least Squares Support Vector Regression Machines (LSSVR). An adaptive recurrent LSSVR machine is proposed and augmented with a Bayesian Inference scheme that allows probabilistic estimates of future health deterioration. Extensions for efficient multi-step long-term prognostics and Remaining Useful Life (RUL) calculation are suggested. Data from a seeded fault test for a UH-60 planetary gearbox plate is used to test the online performance of the prognostics algorithm.

How to Cite

Khawaja, T., & Vachtsevanos, D. G. (2021). A novel Architecture for on-line Failure Prognosis using Probabilistic Least Squares Support Vector Regression Machines. Annual Conference of the PHM Society, 1(1). Retrieved from http://papers.phmsociety.org/index.php/phmconf/article/view/1391
Abstract 188 | PDF Downloads 152

##plugins.themes.bootstrap3.article.details##

Keywords

Bayesian reasoning, data driven prognostics, prognostics

References
Aiwina Heng, Sheng Zhanga, Andy C.C. Tana and Joseph Mathew (2009). Rotating machinery prognostics: State of the art, challenges and opportunities. Mechanical Systems and Signal Processing, Volume 23, Issue 3.
R. Kothamasu, S.H. Huang,W.H. Ver Duin (2006), System health monitoring and prognostics—a review of current paradigms and practices, International Journal of Advanced Manufacturing Technology 28
A.K.S. Jardine, D. Lin, D. Banjevic (2006), A review on machinery diagnostics and prognostics implementing condition-based maintenance, Mechanical Systems and Signal Processing 20
T. Brotherton, G. Jahns, J. Jacobs, D.Wroblewski (2000), Prognosis of faults in gas turbine engines, IEEE Aerospace Conference Proceedings 6
Wang, P., & Vachtsevanos, G. (2003), Fault prognostics using dynamic wavelet neural networks. Artificial Intelligence forEngineering Design and Manufacturing, 15.
Brotherton, T., Jahns, J., Jacobs, J., & Wroblewski, D. (2000), Prognosis of Faults in Gas Turbine Engines. Proceedings of the IEEE Aerospace Conference, IEEE, New York, Vol. 6, pp. 163 –171
Parker, B.E., Jr., Nigro, T.M., Carley, M.P., Barron, R.L., Ward, D.G., Poor, H.V., Rock, D., & DuBois, T.A. (1993), Helicopter gearbox diagnostics and prognostics using vibration signature analysis. Proc. SPIE—The International Society for Optical Engineering, V ol. 1965, pp. 531–542.
Chen, P.-H., C.-J. Lin, and B. Schölkopf (2005), A tutorial on v-support vector machines. Applied Stochastic Models in Business and Industry, 21(2): p. 111-136
Suykens, J.A.K. and J. Vandewalle (1999), Least Squares Support Vector Machine Classifiers, Neural Processing Letters, 9(3): p. 293-300
Gestel, T.V., et al. (2002), Bayesian Framework for Least- Squares Support Vector Machine Classifiers, Gaussian Processes, and Kernel Fisher Discriminant Analysis. Neural Computation 14(5): p. 1115–1147.
J.A.K., S. and V. J. (2000), Least Squares Support V ector Machines for Classification and Non Linear Modeling, in Neural Network world.. p. 20-48.
Van Gestel, T., et al. (2001), Bayesian interpretation of least squares support vector machines for financial time series prediction, in Proc. of the 5th World Multiconference on Systemics, Cybernetics and Informatics. p. 254-259.
Cooper G.F.(1990), The computational complexity of probabilistic inference using Bayesian belief networks, Artificial Intelligence 42 (2–3) 393– 405.
Doucet, A. and X. Wang (2005), Monte Carlo methods for signal processing: a review in the statistical signal processing context. Signal Processing Magazine, IEEE,. 22(6): p. 152 - 170.
Saxena, A., B. Wu, and G. Vachtsevanos (2005), A methodology for analyzing vibration data from planetary gear systems using complex Morlet wavelets, in American Control Conference, 2005. Proceedings of the 2005. p. 4730 - 4735.
Nillson, N.J. (1965), Learning Machines: Foundations of Trainable Pattern-Classifying Systems. New York: McGraw-Hill.
Aizerman, M.A., E.M. Braverman, and L.I.R. er. (1964), Theoretical foundations of the potential function method in pattern recognition and learning, in Automation and Remote Control.. p. 821--837.
Parzen E. (1962). On estimation of a probability density function and mode, Ann. Math. Stat. 33, pp. 1065–1076.
Wu, B., et al. (2004), Data Analysis and Feature Selection for Fault Diagnosis of Helicopter Planetary Gears, in IEEE Autotestcon.
Vachtsevanos, G. et. al. (2006), Intelligent Fault Diagnosis and Failure Prognosis for Engineering Systems, New Jersey: John Wiley & Sons, Inc.
Turlach, B.A. (1993), "Bandwidth selection in kernel density estimation: A review", Discussion Paper 9317, Institut de Statistique, Voie du Roman.
Section
Poster Presentations