The ORSIG hosted a webinar which introduced a project developing a cost-effective condition monitoring system for Floating Offshore Wind Turbine (FOWT) mooring lines for the Carbon Trust’s Floating Wind Technology Acceleration Competition. The presentation was given by the lead consultant for Fugro’s subsea wellhead monitoring service, Stuart Killbourn. In this project the consortium - led by Fugro, AS Mosley and the University of Strathclyde - developed a condition monitoring system for mooring lines of floating wind turbine assets that tracks fatigue and can also be used to detect the occurrence of various events or failure modes. Using such a system is intended to reduce or eliminate visual inspection by ROVs or divers thereby realising a significant cost saving.
A recording of the webinar is available to to watch on demand on IMarEST TV here.
Each floating structure is fitted with instrumentation to record motion and position. The motions are recorded using an inertial sensor package consisting of accelerometers and gyroscopes recording all six degrees of freedom. Position would use a differential GPS receiver. The advantage of using these instruments is that they are easy to attach to the structure and can be located inside the structure, which protects them from weather. The motion data will record the dynamic response of the structure to wave and wind loading typically in the frequency range 0.05 Hz to 5 Hz. The motion signals would first be used to ensure that the dynamic response was matched to the predictions of the simulation model. However, the primary outcome is a knowledge of the stress cycles applied to each mooring line. A knowledge of the stress cycles determines damage accumulation in the mooring lines from fatigue. Advanced modelling techniques of these damage scenarios was implemented to track the condition of various subsea components. What physical inspection of the mooring lines that proves necessary will be guided by the condition indicated as output from the instrumentation system and subsequent data processing. The objective is to perform site inspection only as often as necessary. This will reduce costs and exposure to safety risks. It is envisaged that a continuously monitored structure, where condition is actively assessed, will require far fewer inspections than a structure with no fatigue tracking. The motion and position data will be interpreted by a simulation model of the floating structure. The model can be tailored to the as-built geometry of each of its mooring lines obtained via a post installation seabed survey. Based on the observed motions and positions of the floating structure, the tension cycles in each mooring line is calculated. This information is key to assessing the accumulated fatigue damage sustained throughout the life of the asset.
- How is monitoring of floating turbines currently achieved?
All current floating wind projects are demonstration or pre-commercial. For Hywind Scotland the monitoring consists of GPS, 6DoF motion reference unit and subsea load cells on the six bridles. ORE-Catapult have published data and a description of the system.
- Semi-diurnal tides may not cause many fatigue cycles, but a fast current would cause VIV. Are sites with these fast currents likely to be avoided?
I would agree, VIV of the mooring lines is something that needs to be considered. Monitoring of the motion of the floating hull as per our project, perhaps, would be unlikely to detect VIV occurring. That said, FOWTs are more suited to water too deep to have fixed foundations - so these are naturally further from shore with generally slower tidal flows.
- Have you modelled both fixed mooring/turn-table or is fixed the only option for these floaters as this may help prevent fatigue cracking at the mooring point?
We only modelled the Hywind spar design which is fixed. I am not aware of any reason why other designs could not be simulated - they would need have some kind of modelling during the design phase! Provided the model gives an accurate response, the method would equally apply.
- When looking to obtain an accurate altitude measurement was Radar Altimeter/Radio Altimeter considered as an alternative to high accuracy GPS?
I think you are referring to a wave radar which could track the water surface relative to the sensor installed someway up the tower. This would not be very helpful because the readings would contain the wave undulations and the heave of the FOWT making it difficult to interpret. There are two applications to altitude: (1) tracking the FOWT heave with tide for anomaly detection, and (2) tracking low frequency hydrodynamic modes of the FOWT that the accelerometers struggle to capture.
- Is your process adaptable to semi-submersible platforms? Can it be scaled up to turbines of 15 to 20 MW capacity?
Yes, we believe so.
- If the readings from tension-meters on the moorings offshore, are the same as tensions from modelling; what other value we get by installing this system?
Firstly, our method is a cost-effective alternative to subsea load cells to measure mooring line tension. Any subsea instrumentation presents difficulties in recovering data and providing power - as well as being very difficult to maintain. While our project focussed on tracking fatigue in the mooring lines, similar techniques can be ready applied to the rest of the structure - like the tower section.
- Would the same monitoring be applicable on the subsea power cables to prevent fatigue over pull and issues with MBR?
We did not consider the power cable in our project, but the same methods could be applied to it - all from measurements within the floating hull. I am aware that some strain or bending monitoring is considered by means of a fibre optic core within the cable layout. I'm sure that would be useful if it is cost-effective. It is, however, virtually impossible to maintain.
- Do the wave heights in the 2-6m range contribute the greatest fatigue damage or is this driven by the larger wave heights?
We used the transfer function (TF) generated for a Hs=6m sea state to process simulation data from a range of wave heights. This worked well for Hs between 4m and 8m - perhaps a little wider. In practice, we would have a TF for Hs=9m and use that when waves were higher - and another TF for Hs=12m. Thus, we can accurately capture fatigue in all sea states.
- What degree excess weight of soft/hard marine on the structure/chain/wire has been modelled for potential over stress or drag on the mooring system over time?
We did not include marine growth in the modelling, but I agree that it is one thing that would be good to consider. It would be fairly simple to do. The indications we determined were that wave loading via drag on the mooring lines was relatively small compared to the loading on the hull. Different designs of FOWT might come to different conclusions on this - the spar design is quite a large diameter at the waterline.