A Multi-Fidelity Approach to Predicting Rotor Aerodynamic Interactions

Recent years have seen a huge interest in large electric multicopters for Urban Air Mobility,
commercial package delivery, and military/law enforcement applications. This has lead to
a growing interest in modeling the interactional aerodynamic effects of rotors operating in
close proximity and their impact on performance. While CFD can be used to understand the
detailed physics of multi-rotor interaction, in most cases it is too computationally demanding
for performing optimization and uncertainty quantification studies over a range of parameters.
On the other hand, lower-fidelity models approximate the underlying physics and are computationally
inexpensive, however they are often inaccurate in predicting the desired fields of
interest. With this in mind, in this study we present a multi-fidelity approach that inherits the
computational efficiency of a low-fidelity model while retaining the accuracy of a high-fidelity
method. This approach is based on using the low-fidelity model to span the entire space of
parameters, using these results to identify the key parameter values to perform high-fidelity
simulations, and then using these in a lifting procedure to determine multi-fidelity solutions
at other parameter values. In this manuscript we apply this strategy to determine the disk
plots of lift and drag for the two rotors of a multicopter as a function of the longitudinal and
vertical distance between them. We note that this strategy can substantially improve upon the
low-fidelity results.