Engineering design codes specify a variety of different relationships to quantify vertical variations in wind speed, gust factor and turbulence intensity. These are required to support applications including assessment of wind resource, operability and engineering design. Differences between the available relationships lead to undesirable uncertainty in all stages of an offshore wind project.
Reducing these uncertainties will become increasingly important as wind energy is harnessed in deeper waters and at lower costs. Installation of a traditional met mast is not an option in deep water. Reliable measurement of the local wind, gust and turbulence profiles from floating LiDAR can be challenging. Fortunately, alternative data sources can provide improved characterisation of winds at offshore locations.
Numerical modelling of wind in the lower few hundred metres of the atmosphere is generally much simpler at remote deepwater locations than over complex onshore terrain. The sophistication, resolution and reliability of such models is advancing rapidly. Mesoscale models can now allow nesting of large scale conditions to horizontal scales less than one kilometre.
Models can also provide many decades of wind data, a major advantage over the site specific measurements gathered to support a wind energy development. Model data are also immediately available at the start of a project at relatively low cost.
At offshore locations these models can be validated and calibrated, just above the sea surface, using well established satellite wind products. Reliable long term statistics of near surface wind can be used to quantify winds at the higher elevations applicable to wind turbines using the wide range of existing standard profile relationships.
Reduced uncertainty in these profile relationships will be of considerable benefit to the wider use of satellite and model data sources in the wind energy industry. This paper describes a new assessment of various industry standard wind profile relationships, using a range of available met mast datasets and numerical models.
