Exploring the use of machine learning to support anchor pile design at unseen seabed locations
Exploring the use of machine learning to support anchor pile design at unseen seabed locations
This study investigates the potential role of machine learning (ML) to support design of anchor piles under axial tension, specifically for the case where location-specific data is not available. The analysis is undertaken for an existing site on Australia's North West Shelf where CPT data does exist, with results generated using standard design approaches compared to those based on soil strength predicted from ML. The results demonstrate that ML approaches can generate reasonable design predictions while also providing quantitative assessment of anchor reliability, potentially offering a practical method to advance offshore foundation design in the absence of location-specific data.
Q. Liu; Michael Bertolacci; M. Fraser Bransby; Tim French; Phil Watson
5th International Symposium on Frontiers in Offshore Geotechnics (ISFOG2025)
5 - Data Analytics and Machine Learning
https://doi.org/10.53243/ISFOG2025-240
Contact Information
- Secretary General - Dr A M McNamara
- Email: [email protected]
- Phone: +44 20 7040 8154
- City, University of London
- Northampton Square
- London EC1V 0HB. United Kingdom