Auto-control model building using machine learning regression for extreme response prediction
Darrell Leong A B and Anand Bahuguni AA Lloyd’s Register, 1 Fusionopolis Place, #09-11 Galaxis, 138522 Singapore.
B Corresponding author. Email: Darrell@u.nus.edu
The APPEA Journal 60(1) 155-162 https://doi.org/10.1071/AJ19239
Submitted: 2 December 2019 Accepted: 30 January 2020 Published: 15 May 2020
Abstract
The long-term forecast of extreme response presents a daunting practical problem for offshore structures. These installations are subject to varying sea conditions, which amplify the need to account for the uncertainties of wave heights and periods across a given sea state. Analysis of each sea state involves numerically intensive non-linear dynamic analysis, leading to massive computational expense across the environmental scatter diagram. Recent research has proposed several effective solutions to predict long-term extreme responses, but not without drawbacks, such as the limitation to specific failure locations and the absence of error estimates. This paper explores the practical implementation of control variates as an efficiency enhancing post-processing technique. The model building framework exhibits the advantage of being fully defined from existing simulation results, without the need for external inputs to set up a control experiment. A composite machine learning regression model is developed and investigated for performance in correlating against Monte Carlo data. The sampling methodology presented possesses a crucial advantage of being independent of failure characteristics, allowing for the concurrent extreme response analyses of multiple components across the global structure without the need for re-analysis. The approach is applied on a simulated floating production storage and offloading unit in a site located in the hurricane-prone Gulf of Mexico, vulnerable to heavy-tailed extreme load events.
Keywords: control variates, extreme value, Monte Carlo, mooring analysis, variance reduction.
Darrell Leong is a data scientist developing predictive machine learning models for marine and offshore applications. He completed his PhD on advanced computational mathematics and applied statistics, with significant novel contributions in numerical probabilistic evaluations, uncertainty modelling and simulating black swan events on offshore structures. He is passionate about quantitative research on risk-based decision making, predictive modelling and reliability engineering. |
Anand Bahuguni is a Digital Solutions Lead and Program Manager for Data Analytics at Lloyds Register Singapore. He has over 15 years of experience in Computational Methods, Applied Mathematics and Data Science, coupled with SME knowledge in offshore oil and gas, renewables, process systems and consumer goods sectors. He has published more than 30 conference/journal papers and 10 patents in various disciplines. |
References
American Bureau of Shipping (2016). Guidance notes on selecting design wave by long term stochastic method. (ABS: Houston, TX, USA.)Chai, W., Naess, A., and Leira, B. J. (2018). Long-term extreme response and reliability of a vessel rolling in random beam seas. Journal of Offshore Mechanics and Arctic Engineering 140, 011601-9..
| Long-term extreme response and reliability of a vessel rolling in random beam seas.Crossref | GoogleScholarGoogle Scholar |
de Berg, M., Cheong, O., van Kreveld, M., Overmars, M. (2008). Delaunay triangulations. Chapter 9. In ‘Computational Geometry: Algorithms and Applications’. pp. 191–218. (Springer-Verlag Berlin Heidelberg: Berlin, Germany.)
DNV (1992). Classification note 30.6 (CN-30.6): Structural reliability analysis of marine structures. (Det Norske Veritas: Høvik, Norway.)
DNV (2014). Recommended practice C205 (RP-C205): Environmental conditions and environmental loads. (Det Norske Veritas: H� Norway.)
Glasserman, P. (2003). Variance reduction techniques. In ‘Monte Carlo Methods in Financial Engineering’. (Ed P. Glasserman.) pp. 185–279. (Springer Science+Business Media: New York, NY, USA.)
Leong, D., Low, Y. M., and Kim, Y. (2018). Long-term extreme response prediction of mooring lines using subset simulation. In ‘Proceedings of the 37th International Conference on Ocean, Offshore and Arctic Engineering, Madrid, 17 June 2018’. Paper No.: V003T02A044. (ASME: New York, NY, USA.)
Leong, D., Low, Y. M., and Kim, Y. (2019). An efficient system reliability approach against mooring overload failures. In ‘Proceedings of the 38th International Conference on Ocean, Offshore and Arctic Engineering, Glasgow, 9 June 2019’. Paper No.: V003T02A057. (ASME: New York, NY, USA.)
Low, Y. M. (2016). A variance reduction technique for long-term fatigue analysis of offshore structures using Monte Carlo simulation. Engineering Structures 128, 283–295.
| A variance reduction technique for long-term fatigue analysis of offshore structures using Monte Carlo simulation.Crossref | GoogleScholarGoogle Scholar |
Low, Y. M., and Huang, X. (2017). Long-term extreme response analysis of offshore structures by combining importance sampling with subset simulation. Structural Safety 69, 79–95.
| Long-term extreme response analysis of offshore structures by combining importance sampling with subset simulation.Crossref | GoogleScholarGoogle Scholar |
Nelson, B. L. (1990). Control variate remedies. Operations Research 38, 974–992.
| Control variate remedies.Crossref | GoogleScholarGoogle Scholar |
Ochi, M. K., and Whalen, J. E. (1980). Prediction of the severest significant wave height. Coastal Engineering 1980, 587–599.
| Prediction of the severest significant wave height.Crossref | GoogleScholarGoogle Scholar |
Sagrilo, L. V. S., Naess, A., and Doria, A. S. (2011). On the long-term response of marine structures. Applied Ocean Research 33, 208–214.
| On the long-term response of marine structures.Crossref | GoogleScholarGoogle Scholar |
Tocher, K. D. (1963). ‘The art of simulation.’ (The English Universities Press: London, UK.)
Vázquez-Hernández, A., Ellwanger, G., and Sagrilo, L. V. S. (2011). Long-term response analysis of FPSO mooring systems. Applied Ocean Research 33, 375–383.
| Long-term response analysis of FPSO mooring systems.Crossref | GoogleScholarGoogle Scholar |
