The adoption of propellers is the oldest propulsive solution in aviation, and remains nowadays the preferred one in several segments of the market, being the greenest and most fuel-efficient choice for low-to-mid subsonic speeds. New paradigms of mobility envision the use of propeller powered aircrafts to provide an answer to the issue of future mass transportation in and between major cities.Common aircraft design procedures determine optimal wing design parameters pursuing low drag and weight solutions, but neglecting the influence that propellers' slipstreams have on lift and drag distributions, especially when mounted in tractor configurations.Considering such strong interest in propeller propulsion, and the apparent lack of insightful research in the multi-disciplinary design optimization (MDO) opportunities that propeller-wing integration brings along, the present Thesis was designed with the objective of building a surrogate-based MDO framework for a general aviation aircraft featuring wing mounted tractor propellers. Such objective was achieved by developing appropriate analysis tools to model the most relevant effects of wing-propellers interaction, and by integrating them in suitable optimization architectures enhanced by the adoption of surrogate models. Two research hypotheses underlay such approach, the first claiming that accounting for wing-propeller interaction effects could lead to better designs, and the second claiming that the adoption of surrogate modeling techniques could greatly improve overall optimization performances. The successful accomplishment of the described objective was demonstrated by applying the optimization on an existing aircraft and achieving sensitive fuel savings. The mentioned hypotheses were instead proven under several aspects through the multiple surrogate-based optimization frameworks adopted and the obtained optimal designs. The work proposed as well innovative solutions to perform a full interaction wing-propellers analysis, such as routines to assess and model the propeller streamtube deflection and the wing swirl recovery effects. It also provided guidance for the implementation of surrogate-based optimizations in problems featuring high numbers of design variables and multiple complex non-linear constraints, testing different surrogate models, as well as different frameworks, to achieve local and global optimal designs.iii
ACKNOWLEDGEMENTSThe path leading to the achievement of my Master in Aerospace Engineering has been a challenging and rewarding one, along which all my professors and colleagues at TU Delft have been a huge source of motivation and improvement.This Thesis is the culmination of such journey, and would not have been possible without the support of the people I have been working with over the past year. I would first like to thank my supervisor Dr. Elham of TU Delft for his availability and precious contribution in several aspects of the work. A great thanks is due as well to Professor Brian German who hosted me at the Georgia Institute o...