WindESCo announced that DNV-GL has issued a Statement of Approval for their proprietary method used to quantify increase in power output from their WeBoost solution.
WindESCo’s WeBoost solution combines engineering and machine learning to increase wind farm output without physical upgrades to the wind turbine. One of the key features of WeBoost is quantifying increase in plant output. Through an year-long development effort, the WindESCo technical team developed a robust method to accurately measure the increase in Annual Energy Production (AEP), following implementation of WeBoost recommendations on a wind farm.
Following an extensive review of the methodology and documentation, WindESCo received a Statement of Approval for this method from DNV-GL – a major milestone in achieving widespread acceptance of the approach and providing current and potential customers confidence that the analysis is fair to both parties. Relying exclusively on 10-minute SCADA statistics from the turbines of interest and appropriate control turbines, WindESCo’s power comparison method has been successfully employed to measure the AEP increase and corresponding uncertainty for completed WeBoost contracts. Results show good reliability for measuring energy production improvement gains across many different turbine OEMs and wind farms sizes.
In traditional upgrade performance measurement campaigns, power improvement is measured on one or two example turbines, often using a meteorological tower to make wind speed measurements. The results are then extrapolated to the entire wind farm. This requires additional instrumentation, results in relatively high uncertainties, and does not lend itself to analysis of individualized turbine optimization as is provided by WeBoost. In other cases, nacelle wind speed measurements are used; however, this introduces additional uncertainties that are hard to quantify and does not work well when the optimization introduced may influence the wind speed measurement. In contrast, WindESCo’s comparison method does not rely on wind speed measurements and is applied to all optimized turbines on the windfarm. This enables improvement measurement of individual turbine specific adjustments and reduces the overall uncertainty by making the entire windfarm part of the measurement population.