A Data-driven Bayesian Ascent Method for Maximizing Wind Farm Power Production

J. PARK, S. KWON, K.H. LAW

Abstract


This paper discusses a data-driven, cooperative strategy to maximize wind farm power production. By strategically coordinating the control actions of the wind turbines to actively mitigate the wake interference, the total wind farm power production can be improved for a given wind condition. To determine the optimum coordinated control actions of the wind turbines using only power measurements collected from the wind turbines, we employ the Bayesian Ascent (BA) method, a probabilistic data-driven optimization scheme. Wind tunnel experiments using 4 scaled wind turbine models have been conducted to validate (1) the effectiveness of the cooperative control strategy and (2) the efficiency of the BA algorithm in determining the optimum control actions of the wind turbines using only the power measurement data.

doi: 10.12783/SHM2015/280


Full Text:

 Subscribers Only