A data-driven approach for identification and compensation of wind turbine inherent yaw misalignment

Abstract

The yaw control system plays a significant role in the power generation performance of wind turbines. As one of the main components applied on controlling the wind turbine nacelle position in parallel with the inflow wind, it directly determines the maximum available wind energy to capture. However, the inherent misalignments on yaw error caused by improper calibration of the wind vanes during wind turbine installation, are usually severe and impact on the performance of yaw control strategies. In this study, a data-driven inherent misalignment identification and compensation scheme for yaw error is proposed. The real-time data from turbines are collected and a power generation curve is located for each yaw error interval. Subsequently, multiple power curves are evaluated via output performance indicator analysis for determining the actual inherent misalignment value. The entire system is implemented and testified with the SCADA data collected from real wind turbines.