Maximizing Wind Farm Power: Wake Control Using Machine Learning Algorithms
Date
2020-08
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Windfarms are designed with turbines in proximity due to land and transmission line constraints. As a result, depending on the wind direction, the wake from upstream turbines will impact the downstream turbines, reducing their power generation and the overall wind farm power generation. To mitigate this effect, several wake control technologies have been shown benefit. One strategy to control the wake is to curtail the power of the most upstream turbine. In this research, machine learning algorithms are applied to model the wake effect using Power, nacelle direction, windspeed, and RPM. Next, the model is used to investigate the potential for increasing the power in the downstream turbine by derating the upstream turbine. Although there is a potential for increasing the power, the model performance was not as expected due to the insufficiency of actual data that represents the performance of the downwind turbine when the upstream turbine is curtailed.
Description
Keywords
Citation
Salek, B. (2020). Maximizing Wind Farm Power: Wake Control Using Machine Learning Algorithms (Unpublished master's project). University of Calgary, Calgary, AB.