The sea state can make working offshore, particularly during lifting operations, a tricky business. Peter Naaijen, ship and offshore hydromechanics expert at Delft University of Technology, explains to Mark Brierley how current research is aiming to predict wave-induced vessel motion in real time.

PN

Peter Naaijen has worked as assistant professor at the Delft University of Technology since 2003. His main research fields have been passing ship induced waves in confined waters, wind-assisted propulsion for commercial shipping and wave-induced motion prediction using remote observations from nautical radar.

Even though controlling the weather may still be something only Wilhelm Reich can dream of, better predictive technology could be a boon for those working offshore, allowing them to make much better use of windows of fair weather. Cloudbusters need not apply.

E&P processes have evolved a great deal over the decades, overcoming, or at least delaying, the decline of mature reserves, and keeping pace with the world’s insatiable appetite for fossil fuels. Technological advances have turned the industry over the last century from ‘drill and hope’ to one of the most efficient extractive industries on the planet. Technology can only do so much, though. One of the biggest inefficiencies working offshore that is still beyond the taming influence of technology is the weather. Less-than-ideal conditions can severely hamper much of the work carried out by floating vessels. In the UK, adverse weather can limit work on floating vessels 65km or so from shore to 200 days in an average year.

Motion on the ocean

Specifically, waves can cause a number of problems for vessels working at sea.

"Think of the installation of PLEM/PLETs, where, due to crane vessel motions, the impact load due to touch down on the seabed can cause damage," explains Peter Naaijen, from Delft University of Technology, who is currently undertaking research into the prediction of wave-induced vessel motion. "Offloading operations from floating craft where, after lift-off, recollision of the load with the vessel is a risk. Helicopters landing on floating vessels is another example."

To minimise the chances of any of these eventualities, the university is currently co-funding the ‘PROMISED Operations’ (prediction of wave-induced motions and forces in ship, offshore and dredging operations) research project, along with the Dutch Ministry of Economic Affairs, Agriculture and Innovation, the University of Twente, Maritime Research Institute Netherlands, Ocean Waves, Allseas, Heerema Marine Contractors and IHC.

Operations at sea are currently dictated by a number of statistical parameters, such as wave height and probability of impact loads. If any of these fall outside of defined parameters, workers must down tools until the weather improves, but Naaijen argues that this approach wastes time.

"From a statistical standpoint, even within conditions that are considered unworkable, there may be windows of opportunity in which vessel motion is below critical values," he contends. "Being able to predict these quiescent periods in which vessel motion is at a minimum makes it is possible to time the critical part of an operation. We call this a shift from a statistical to a deterministic approach to operability."

This allows small windows of opportunity for work to be carried out, even in unfavourable conditions, saving time and money for operators. Working in such tight timeframes requires extremely accurate prediction of vessel motion, though.

Swell guy

X-band nautical radar makes it possible to observe waves 1,500-2,000m away. Although designed to track other vessels in the area, it is possible to derive wave patterns from the unfiltered images. These can then be used to initialise a wave-propagation model and predict lulls in swell, and, therefore, in vessel motion.

"One big inefficiency working offshore that can’t be tamed by technology is weather."

"A prediction of the waves arriving at the vessel is obtained anywhere from tens of seconds to three minutes in advance, depending on conditions and radar range. Conventional methods are used to translate wave forecasts into ship motion predictions and pre-calculated responses of the vessel to the conditions are used."

The leap from seconds-in-advance prediction to real-time prediction may seem like a small one, but analysing radar images at such a rate can be a cumbersome endeavour. "This is the bottleneck of the whole prediction system," confirms Naaijen. "Techniques that use nautical radar to observe waves mainly aim to detect surface current and water depth – both of which can be derived from wave patterns – and eventually directional energy density spectra of the waves. In the current research project, we use improved methods that are more suitable for detecting the individual waves, thus enabling more accurate, real-time prediction."

Going beyond conventional technology has allowed this final hurdle to be overcome, but the work doesn’t stop there.

"More accurate observation systems are likely to emerge in future, which is the most important ingredient for enhanced accuracy of prediction" says Naaijen.

Just improving the technology behind prediction alone is not enough though: "I believe a lot can be gained in efficient communication and presentation of predictions to the users," he continues.

"Sharing the exact relevant information with a helicopter pilot, or between different vessels involved in an operation, is crucial for successful use of prediction systems."

It must be an effective marriage of users and technology to weather the storm.


Probing the depths

In 2009, Naaijen took part in the OWME (onboard wave and motion estimation) industry project, along with colleagues from the Delft University of Technology and MARIN, which used Ocean Waves’ WAMOS II radar image processing software to provide real-time time traces of wave elevation. The project aimed to predict wave elevation and vessel motions 60-120 seconds ahead, using a number of scale model tests in MARIN’s research centre in the Netherlands, and to discover the best combination of prediction model parameters and probe set-ups for various sea states. It was found that the position of probes and prediction location had a significant impact on the accuracy of prediction and informed future research.