Operational Modal Analysis of Rotating Structures Using DIC: Rigid Body Motion Compensation and Image Reordering

Digital Image Correlation (DIC) is a non-contact, full-field technique that provides a reliable approach for studying the modal parameters of rotating structures. When combined with Operational Modal Analysis (OMA), it enables the identification of these parameters directly from the frequency content of the structural response, without requiring knowledge of the applied force. This makes DIC and OMA particularly valuable for analyzing rotating components, such as turbine and compressor bladed disks in aeronautical engines, under operational conditions. Despite its effectiveness, applying DIC to rotating structures presents unique challenges. Traditional DIC algorithms require small inter-frame structural motions, making incremental DIC necessary for analyzing rotating structures. However, this approach accumulates errors over multiple rotations, leading to displacement divergence over time. Additionally, because both rotational and vibrational motions are captured by the cameras, compensation methods must be applied to correct displacement results for rigid body motion. This thesis addresses these challenges by employing existing rigid body motion compensation algorithms and proposing a method to mitigate error divergence by reordering images based on the structure’s angular position rather than acquisition time. This method effectively simulates a single revolution, significantly reducing cumulative errors. The validity of this approach is demonstrated through an experimental analysis on a dummy bladed disk. First, impact testing is performed to identify the specimen’s modal parameters, with displacements measured using DIC and analyzed through Experimental Modal Analysis (EMA). The results are then validated with a Finite Element Model. Next, a high-speed imaging setup captures the rotating bladed disk at different angular velocities while the blades are excited by an impulsive magnetic force. These images are processed using an incremental DIC approach to extract full-field displacement time histories. These data, corrected using rigid body motion compensation algorithms, are then analyzed using Operational Modal Analysis (OMA). A comparison between time and angularly sorted images shows a significant reduction in displacement divergence, confirming the effectiveness of the proposed methodology in minimizing cumulative errors.