Offshore structures accumulate damage during operation due to the fluctuating environmental and operational forces, thus, these structure are prone to fatigue failure. The design of offshore structures avoids this failure by simplifying reality and setting up a designed lifetime for save operation.
Many offshore structures in the North Sea are, however, reaching the end of their designed lifetime, but the actual integrity of the structures is unknown. The fatigue assessment of existing offshore structures requires sensors at all fatigue critical locations of the structure. Unfortunately, these locations are often inaccessible and harmful to the sensors due to the hostile environment of the North Sea. This project aims to estimate the stress history of offshore structures during operation by virtual sensing. This will give us a better understanding of the remaining lifetime of the existing structures in the North Sea.
The North Sea is a hostile environment and it will likely damage any sensors subsea. Therefore, direct measurements of the stress history is near impossible for offshore platforms. Using virtual sensing, however, we utilize sensors above water to estimate the stress history in unmeasured points [1].
A virtual sensor is an alternative to a physical sensor directly measuring a desired quantity of a system. Three components are needed to create virtual sensors and they are termed: system model, physical sensors, and process model.
Operational Modal Analysis is used to extract physical information (modal parameter) from the structures during operation. This is based on the random vibration measured from the top of the platform.
Exact information from the ambient vibrations is vital to get good estimations of the stresses. Therefore, research is conducted to reduce the statistical errors in Operational Modal Analysis.
M. Tarpø, T. Friis, C. Georgakis, R. Brincker, The statistical errors in the estimated correlation function matrix for operational modal analysis, Journal of Sound and Vibration, 466 (2020), 115013
