Prediction of load distribution based on DPIV images for a flexible wing undergoing plunge motion

Abstract

In this thesis, main purpose is the prediction of force and pressure distribution on a wing using image processing. Determining the pressure distribution is the main objective for investigation aerodynamical properties. Two different wing types as Lexan and 3DPrinted, are tested. An image processing code is created to determine the pressure distribution with coding software MATLAB by MathWorks as a version of R2021a. Image processing is used for determining the deformation and displacement of wing. Force distribution is calculated with outputs from image processing. Image data sets and force measurement results are obtained from a previous study performed in Trisonic Laboratory of Istanbul Technical University. Main objective of the study is validation and optimization of an image processing method which is combined with experimental studies for determining pressure distribution and efficiency from deformation of a wing. Research of this thesis is related to the PhD study of Son (2018), titled "Experimental Investigation of Flexibility Effects in Flapping Wing Aerodynamics". Four different models are used in the experiments which are Rigid, Lexan, 3D Printed, whose material is colorfabb_xt transparent, and Acetate. Image processing is applied only to 3D Printed flexible wing cases. Acetate type is eliminated due to thickness of plate whose thickness is close to glass sphere particle size and eliminated in image processing. Lexan type wing is eliminated due to the brightness issue of the image data set. Material properties and natural frequencies are calculated by weight deflection method. Averaged damped natural frequencies are calculated as 10.75 rad/sec for Printed and 14.26 rad/sec for Lexan and averaged damping ratios as 0.305 and 0.175 respectively. Added masses are 1.133 kg for both plates. In the post processing of images, 9th degree polynomial curve fitted into binarized image. Positions and deflection of nodes are calculated. Forces are calculated for nodes and load distribution is determined. Pressure distribution of wings are calculated for 5 different frequencies with 4 different amplitudes and compared with the integrated value of experiment. A reasonable agreement is found between measured force in experiments and forces determined from image processing for some of the cases.