The advent of auctions across the globe has resulted in a considerably lower price of electricity. As assets are ageing, owners also need to look at rising O&M costs to maintain profitability. These changes are making asset developer revisit their bidding strategy and also relook their O&M contracts to reduce financial risks.
One key metric used in O&M contracts pertains to availability, generally referring to the ability of the O&M operator to keep the wind project working and producing electricity. The O&M contract defines an availability percentage across different months which needs to be adhered to avoid the penalty clause from kicking in.
The availability metric provides insights on turbine performance and is a key metric tracked by the asset owner. Currently, availability is defined in terms of either time-based or production based. Time-based availability takes into consideration the time for which the turbine was available for generation in a period. Whereas, production-based availability takes into consideration the potential energy that can be generated by an asset under site conditions. Most O&M contracts focus on time-based availability as the key metric. The key area in such a case for the asset owner is to focus on how this availability is being calculated and ensure all cases pertaining to low wind, high wind, site conditions during turbine operations are considered. These considerations will help asset owners get a true understanding of the availability of the wind farm.
However, we believe production-based availability should be the key metric for asset owners. There are quite a few ways to compute potential energy for production-based availability, including the use of
Other computational methods can also be used in lines with IEC standard 61400-26-2. The goal behind using production-based availability is to provide a truer assessment of wind farm performance and identify under-performing wind turbines. Although computation of production-based availability was a challenge earlier for a reasonably sized wind farm, the advent of data analytics has made it easier to store data and setup computational rules.
I will illustrate with an example on how production-based availability provides a better assessment than time-based. Consider the power curve of the turbine below in reference to the power curve. This power curve is for the month of July. The turbine can be seen in blue whereas the reference power curve can be seen in red.
The turbine had a time-based availability of 98.3%, which is above the contractual obligation whereas the production-based availability was 94.5%, computed using nacelle anemometer wind speed. This can be construed as an extreme case but our analysis across different wind farms of our customers provided us with similar insights. We conducted this analysis across different months and found the following results
In conclusion, we are witnessing some asset owners make a switch to production-based availability and with the advent of data analytics, we believe this implementation becoming a standard in a few years. We at Algo Engines are helping asset owners track availability and identifying under-performing turbines to further enhance plant generation..