Abstract
It has long been known that the standard equation—Morisons equation—for the prediction of fluid loading forces on slender members, is inadequate outside a fairly narrow regime of wave conditions. There have been many attempts to improve on Morisons equation over the years, including a number based on nonlinear system identification. Some years ago, the current first author, together with collaborators, proposed an identification methodology based on polynomial NARMAX/NARX models. The objective of the current paper is to update that methodology, taking into account modern practice in machine learning. In particular, an approach based on Gaussian process NARX models will be demonstrated, which has the advantage of bypassing the polynomial structure detection problem and also of providing natural confidence intervals for predictions. The approach will be demonstrated on real data for wave forces in a directional sea. The current paper will also take the opportunity to critically highlight a number of weaknesses of the original study in the light of modern best practice in machine learning.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The acronym NARMAX stands for Nonlinear Auto-Regressive Moving Average with e(X)ogenous inputs. The model accounts for process dynamics via the AR part, and noise dynamics via the MA. If one neglects the noise model by assuming that the noise is white Gaussian, the overall model is denoted NARX.
- 2.
Within other sectors of the system identification community, a different terminology is commonly adopted; it is often the case that computing the OSA predictions is simply referred to as prediction, while computing the MPO predictions is referred to as simulation. This is a little unfortunate, but the authors will continue with the OSA/MPO terminology to ensure consistency with their previous work.
- 3.
The coefficients of the noise model terms are not included in the table. The noise terms are only included in the model as a means of reducing parameter estimation bias, they are not used in making predictions.
References
Morison, J.R., O’Brien, M.P., Johnson, J.W., Schaf, S.A.: The force exerted by surface waves on piles. Pet. Trans. 189, 149–157 (1950)
Obasaju, E.D., Bearman, P.W., Graham, J.M.R.: A study of forces, circulation and vortex patterns around a circular cylinder in oscillating flow. J. Fluid Mech. 196, 467–494(1988)
Sarpkaya, T.: Resistance in unsteady flow - search for a model. Technical Report, Naval Postgraduate School, Department of Mechanical Engineering, Monterey, CA, 93943 (1978)
Worden, K., Stansby, P.K., Tomlinson, G.R., Billings, S.A.: Identification of nonlinear wave forces. J. Fluids Struct. 8, 19–71 (1994)
Bishop, C.M.: Pattern Recognition and Machine Learning. Springer, Berlin (2013)
Rasmussen, C.E., Williams, C.: Gaussian Processes for Machine Learning. The MIT Press, Cambridge (2006)
Bishop, J.R.: Aspects of large scale wave force experiments and some early results from Christchurch Bay. Technical Report NMI-R57, National Maritime Institute (1979)
Leontaritis, I.J., Billings, S.A.: Input-output parametric models for nonlinear systems, Part I: deterministic nonlinear systems. Int. J. Control 41, 303–328 (1985)
Leontaritis, I.J., Billings, S.A.: Input-output parametric models for nonlinear systems, Part II: stochastic nonlinear systems. Int. J. Control 41, 329–344 (1985)
Billings, S.A.: Nonlinear System Identification: NARMAX, Methods in the Time, Frequency, and Spatio-Temporal Domains. Wiley-Blackwell, Oxford (2013)
Billings, S.A., Jamaluddin, H.B., Chen, S.: Properties of neural networks with applications to modelling non-linear dynamical systems. Int. J. Control 55, 193–1350 (1992)
Chen, S., Billings, S.A., Cowan, C.F.N., Grant, P.M.: Practical identification of NARMAX models using radial basis functions. Int. J. Control 52, 1327–1350 (1990)
Worden, K., Stansby, P.K., Billings, S.A., Tomlinson, G.R.: On wave force analysis using system identification. In: Proceedings of BOSS 92, Imperial College, London, pp. 415–430 (1992)
Schetzen, M.: The Volterra and Wiener Theories of Nonlinear Systems. John Wiley Interscience Publication, New York (1980)
Worden, K., Tomlinson, G.R.: Nonlinearity in Structural Dynamics: Detection, Modelling and Identification. Institute of Physics Press, Bristol (2001)
Murray-Smith, R., Johansen, T.A., Shorten, R.: On transient dynamics, off-equilibrium behaviour and identification in blended multiple model structures. In: Proceedings of European Control Conference, pp. BA–14 (1999)
Kocijan, J.: Dynamic GP models: an overview and recent developments. In: Proceedings of 6th International Conference on Applied Mathematics, Simulation and Modelling, pp. 38–43 (2012)
Worden, K., Manson, G., Cross, E.J.: Higher-order frequency response functions from Gaussian process NARX models. In: Proceedings of 25th International Conference on Noise and Vibration Engineering, Leuven (2012)
Chance, J.E., Worden, K., Tomlinson, G.R.: Frequency domain analysis of NARX neural networks. J. Sound Vib. 213, 915–941 (1997)
Krige, D.G.: A statistical approach to some mine valuations and allied problems at the witwatersrand. Master’s thesis, University of Witwatersrand (1951)
Neal, R.M.: Monte Carlo implementation of Gaussian process models for Bayesian regression and classification. Technical Report arXiv: Physics/9701026 (1997)
Mackay, D.J.C.: Gaussian processes - a replacement for supervised neural networks. lecture notes for tutorial. In: International Conference on Neural Information Processing Systems (1997)
Billings, S.A., Chen, S., Backhouse, R.J.: Identification of linear and nonlinear models of a turbocharged automotive diesel engine. Mech. Syst. Signal Process. 3, 123–142 (1989)
Snelson, E., Ghahramani, Z.: Sparse Gaussian processes using pseudo-inputs. In: Advances in Neural Information Processing Systems. MIT Press, Cambridge (2006)
Quinonero-Candelo, Q., Rasmussen, C.E.: A unifying view of sparse approximation Gaussian process egression. J. Mach. Learn. Res. 6, 1939–1959 (2005)
Giraud, A.: Approximate methods for propagation of uncertainty with Gaussian process models. PhD thesis, University of Glasgow (2004)
McHutchon, A., Rasmussen, C.E.: Gaussian process training with input noise. In: Proceedings of 24th International Conference on Neural Information Processing Systems (2011)
Chen, S., Billings, S.A., Luo, W.: Orthogonal least squares methods and their application to non-linear system identification. Int. J. Control 50, 1873–1896 (1989)
Worden, K.: Confidence bounds for frequency response functions from time series models. Mech. Syst. Signal Process. 12, 559–569 (1998)
Worden, K., Hensman, J.J.: Parameter estimation and model selection for a class of hysteretic systems using Bayesian inference. Mech. Syst. Signal Process. 32, 153–169 (2011)
Wilson A, G., Knowles, D.A., Ghahramani, Z.: Gaussian process regression networks. In: Proceedings of the 29th International Conference on Machine Learning, Edinburgh (2012)
Gramacy, R.B.: tgp: an R package for Bayesian nonstationary, semiparametric nonlinear regression and design by treed Gaussian process models. J. Stat. Softw. 19, 9 (2007)
Lázaro-Gredilla, M., Titsias, M.K.: Variational heteroscedastic Gaussian process regression. In: Proceedings of the 28th International Conference on Machine Learning, Bellevue, WA (2011)
Worden, K., Manson, G., Tomlinson, G.R.: A harmonic probing algorithm for the multi-input Volterra series. J. Sound Vib. 201, 67–84 (1997)
Acknowledgements
The authors would like to thank Ramboll Oil and Gas, Denmark, for financial support for TR; they would also like to specifically thank Dr Ulf Tyge Tygeson for various useful discussions regarding wave loading.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 The Society for Experimental Mechanics, Inc.
About this paper
Cite this paper
Worden, K., Rogers, T., Cross, E.J. (2017). Identification of Nonlinear Wave Forces Using Gaussian Process NARX Models. In: Kerschen, G. (eds) Nonlinear Dynamics, Volume 1. Conference Proceedings of the Society for Experimental Mechanics Series. Springer, Cham. https://doi.org/10.1007/978-3-319-54404-5_22
Download citation
DOI: https://doi.org/10.1007/978-3-319-54404-5_22
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-54403-8
Online ISBN: 978-3-319-54404-5
eBook Packages: EngineeringEngineering (R0)