Non-contact vibration monitoring of rotating wind turbines using a semi-autonomous UAV

摘要

With the recent demands for more efficient clean renewable energy sources, wind turbines are designed that have large rotor blade diameters. These large-sized wind turbines need to be periodically monitored to prevent catastrophic failures. This paper aims to develop a practice for obtaining the vibration characteristics of wind turbine blades that can be eventually used for structural health monitoring of these structures. This monitoring technique needs to be robust and non-contact to prevent any interference with the operation of the wind turbine. In this work, a digital image correlation (D1C) system installed on a drone is used as a sensing technique to obtain the dynamic characteristics of rotating wind turbine blades. The DIC uses a stereo-camera system to record the deflections of the blades and provides non-contact measurements. The unmanned aerial vehicle (UAV) enables on-site robust measurements. Furthermore, a dynamic stitching technique is used after DIC measurement to obtain vibration characteristics of the entire blade with high accuracy. The proposed health monitoring technique can be used by engineers for remote structural health monitoring of wind turbines during operation in both offshore and inland wind farms.