State-of-the-Art and Future Directions for Predictive Modelling of Offshore Structure Dynamics Using Machine Learning

摘要

Ramboll Oil and Gas are leading the field in the development of Structural Health Monitoring Systems (SHMS) for offshore structures. This paper outlines the State-of-the-Art process for predictive maintenance that Ramboll have developed and implemented for offshore structures. This system is one of the first, if not the only one, that creates a maintenance schedule based on knowledge of the structure's current state. The State-of-the-Art methods of today, as adopted by Ramboll, encompass advanced analysis methods ranging from linear and non-linear system identification, expansion processes, Bayesian FEM updating, wave load calibration, quantification of uncertainties from measured data, damage detection and structural re-assessment analysis to Risk- and Reliability-Based Inspection Planning (RBI) analysis. The paper will be the first in a series of papers that will outline various promising methods contributing to an even better understanding of the issues at stake in the offshore structures context.