Multi-Fidelity Bayesian Optimization of a Coaxial Rotor for eVTOL Aircraft

The importance of aerodynamic fidelity in capturing coaxial rotor performance in hover and edgewise flight is considered. Low- and mid-fidelity aerodynamic methods based on momentum and vortex theory are used, and the extent to which the complexity of rotor interactions is appropriately captured by the different models is explored. The results from both model fidelities are deployed within a Bayesian optimization (BO) framework. This framework pairs a multi-fidelity Gaussian process (GP) with an acquisition function to guide the search for an optimal coaxial geometry over hover and forward flight modes. We demonstrate the use of the multi-fidelity Bayesian optimization to capture the behavior of the high-fidelity model over the design space with reduced computational needs. Results of the optimization show axial spacing to be a dominant factor, with different design implications for hover and forward flight. Parameterized chord and twist distributions were also critical. Importantly, a number of design candidates were encountered during the optimization that performed very closely to the global optimum, highlighting the non-convex and multi-modal nature of the problem.