Just a few years ago wind resource assessment was a relatively basic exercise. Studies were based on low-quality measurement data, not always from certified equipment and often with met mast heights that took no account of future hub heights. No long-term correction was applied and simplified flow models were used to estimate the resource beyond the met mast positions. Such limitations at this important phase of wind project development resulted in high uncertainty about production and a potential increase in costs over the lifetime of a project.
It is clear that assessment of the wind regime and the corresponding annual energy production (AEP) are highly important for the viability of a planned wind farm. Any uncertainty, therefore, over the AEP is a crucial factor when financing a project, affecting important decisions such as site selection, operational strategies and, perhaps most importantly, the choice of turbine to be installed.
The field of wind resource assessment has seen significant change over the past few years, with the main driver being to achieve greater certainty over these AEP calculations. From a turbine manufacturer's point of view it is equally important to understand the load regimes under which the turbines are expected to operate so that this knowledge can be fed back and applied to the design and control.
Expensive lessons from the past have resulted in drastic improvements in wind measurement standards, both in terms of mast instrumentation and height, which greatly benefits project planning. Mast heights can now match hub heights - the most common request, according to Wind Measurement International, is for 100m or 120m, and research company Fraunhofer has produced one of 200m. Long-term correction to avoid dependency on a high or a low wind year measured has also become best practice when assessing AEP.
However, knowing the wind at the mast positions is not sufficient. The entire wind field across the site must be explored, which is where wind-flow models come into play. The wind beyond the masts is assessed with mathematical flow models, with input from the mast data and terrain conditions. How well a model performs depends on the wind input, terrain, atmospheric conditions at the site and, crucially, the model itself. And this is where the groundbreaking development in wind resource assessment has taken place.
Limited computing power has, until only about five years ago, forced project planners to make assumptions and to simplify models. Such simplifications inevitably diluted the quality of the turbulence modelling and other complex flow patterns, skewing the results, as well as giving insufficient weight to the true meteorological conditions at the site. Changes in conditions according to time - seasonal and time of day - were also neglected because of the use of time averaging that was often not appropriate.
But there has been immense growth in computing power in the wind industry, through cheaper supercomputers and the use of cloud computing. Equally dramatic advances in software have transformed tools, such as computational fluid dynamics (CFD) models to assess wind flow over terrain, from mere academic notions into successful business operation modelling aids. Major players in the wind industry, alongside academic institutions, are investing in continuous improvement of CFD models to better represent wind flow over complex terrain and to incorporate offshore wind-wave interaction, as well as combining with tools like numerical weather forecast models (mesoscale models) to take account of full atmospheric variability.
The use of CFD combined with actuator disk or actuator line models representing the spinning wind turbines within the CFD domains — for example to study windand wake-induced turbulence — is also finding its way into operational business processes.
Investment in wind flow modelling gives direct benefits as it is the main method of minimising uncertainty in wind resource assessment, significantly affecting the levelised cost of energy through AEP, financing, optimal turbine selection and operational strategies.
A typical misconception is that a wind farm will achieve a higher AEP output if we understand more about the wind and its variability. Better understanding can just as easily lead to a lower AEP as a higher one. A lower AEP means that the P50 (the predicted average annual energy yield, which is 50% likely to happen) will go down, but the relationship between the P50 and the P90 (a yield prediction that is 90% likely to happen) or P99 values will improve, leading to greater certainty for financing decisions. The wind resource assessment might result in a lower AEP — but also a more realistic prediction for investors.
Wind resource assessment of a project is largely carried out as a guide to site development and financing. But a sound understanding of the complete wind regime is equally important when it comes to operational strategies, installation and service costs. It also provides valuable information for turbine design and control, risk management and development of new service products. This underlines how the quality of wind resource assessment affects the entire life cycle of a wind energy project, from design to end of life.
Despite the effort being put into improving wind resource assessment, there remains a problem with awareness of and trust in the newer models, in terms of what is accepted as input to a bankable wind resource assessment report. When negotiating a contract or financing of a wind farm, developers and financers often still prefer to rely on the older, simpler models over the newer, higher-quality yet less well-known approaches.
This is a challenge that the industry has to overcome; the time taken for the new methods to be proven and accepted is too long, and is costing too much.
Effective wind resource assessment is crucial, not only for wind project planning and financing but also to secure quality input over a project's life. Within wind variability lie risks that have to be identified and assessed, but also opportunities for the industry to harvest greater value. The recent exponential increase in computing power has opened the door to advanced mathematical modelling in day-to-day business operations, removing the need for crude assumptions and approximations. Yet the gap between best knowledge and best practice remains huge and costly.
The parties responsible for bringing new science and methods to the market now have a responsibility to engage in and drive best practice to an adequate level.
As the wind industry continues to mature, with ever-increasing numbers of turbines being installed around the globe, the overall level of understanding within this field is growing, as is the business case certainty of wind energy projects. However, this can only continue if lessons from the past are used in the smartest way, enabling findings to be implemented and shared in a timely way through improved measurement methods, modelling tools and best practice.
Line Gulstad is director of plant siting and forecasting, plant solutions, Vestas Technology and service solutions at Vestas