Extreme weather is a constant threat to the service of a wind turbine, as it may cause damage to the rotor blades or even lead to the collapse of a supporting tower. A change in the dominant frequency may indicate a deviation from the design of the stiffness of the turbine-tower system. Dynamic measurements of the lateral displacement of four 2-MW wind turbines of the same model have been obtained using a ground-based microwave interferometer. Vibration analysis, including frequency domain decomposition and reassigned spectrograms, is undertaken to evaluate the frequency contents of a turbine-tower system. Dominant frequency, along with other peak frequencies, is readily identified when the rotor is stopped or operated at low to medium speed. Dynamic characteristics found in the vibration analysis are closely examined to identify indicators of structural deficiency in the supporting tower of a wind turbine. The results are promising as they combine noncontact dynamic measurements and vibration analysis for the condition assessment of wind turbine tower systems. Preliminary numerical results also indicate possible further enhancement of the proposed vibration analysis, using the relation curve that highlights the discrepancy percentage of normalized deflections with respect to the change in stiffness of the supporting tower of a wind turbine.

Digital Image Correlation (DIC) technique provides an accurate and efficient solution for measuring both 2D and 3D displacements of large structures. However, a successful DIC implementation requires unique patterns or markers on the target surface. Creating artificial markers on large structures is a time-consuming and challenging task. Any error while developing or identifying unique markers could lead to unreliable and inaccurate DIC results. This study introduces a novel Artificial Intelligence (AI)-based approach to identify and generate distinctive feature-rich natural markers for DIC. The proposed technique includes a crucial preprocessing step, which comprises an instance segmentation model built with the You Only Look Once (YOLOv8) and the Segment Anything Model (SAM) deep learning algorithms. This model ensures that the markers are integral to the structure rather than a part of the background. Further, the developed methodology employs the KAZE feature-based clustering approach to identify poly-shaped non-intersecting regions as a DIC marker for achieving strong correlation. This study incorporates a wind turbine tower dataset to validate and demonstrate the proposed methodology. The performance of the developed technique is evaluated with respect to the conventional manual marker selection approach and recently developed marker generation methodologies. It is observed that the proposed methodology is 11 times faster and reduces memory consumption by 63%. Moreover, it excludes feature-less regions and can successfully determine the optimal feature-rich DIC marker (in the form of a non-intersecting poly-shaped marker) for achieving strong correlations.

Unmanned aerial vehicle (UAV) can gather the spatial information of huge structures, such as wind turbines, that can be difficult to obtain with traditional approaches. In this paper, the UAV used in the experiments is equipped with high resolution camera and thermal infrared camera. The high resolution camera can provide a series of images with resolution up to 10 Megapixels. Those images can be used to form the 3D model using the digital photogrammetry technique. By comparing the 3D scenes of the same wind turbine at different times, possible displacement of the supporting tower of the wind turbine, caused by ground movement or foundation deterioration may be determined. The recorded thermal images are analyzed by applying the image segmentation methods to the surface temperature distribution. A series of sub-regions are separated by the differences of the surface temperature. The high-resolution optical image and the segmented thermal image are fused such that the surface anomalies are more easily identified for wind turbines.