Multiple Rotor Acoustic Control

Author:

Tumelero Valente, Vitor

Graduate Program:

Aerospace Engineering

Degree:

Doctor of Philosophy

Document Type:

Dissertation

Date of Defense:

March 27, 2024

Committee Members:

Amy Pritchett, Program Head/Chair
Simon Miller, Major Field Member
Eric Greenwood, Outside Field Member
Amy Pritchett, Major Field Member
Eric Johnson, Chair & Dissertation Advisor
Sean Brennan, Outside Unit Member

Keywords:

Synchrophasing
Phase control
Noise control
Acoustic control

Abstract:

Many electric Vertical Take-Off and Landing (eVTOL) vehicles and many small unmanned aircraft systems (sUAS) use multiple rotors. These multirotor aircraft are typical for proposed Advanced Air Mobility (AAM) operations for robust, effective, and automated passenger and cargo transportation within densely populated areas. These vehicles tend to fly much closer to the ground than typical airplanes. Therefore, one of the main barriers to AAM operations is community noise. At the same time, these aircraft often have overactuated flight control schemes, meaning that the aircraft have more controls (e.g., individual rotor RPM or blade pitch) than is necessary to control vehicle forces and moments. This provides mechanisms for noise reducing control strategies, so as to enable the large scale deployment of such vehicles in and around communities. The fundamental noise generation processes in a multirotor eVTOL are similar to other aircraft. Tonal noise can be the dominant noise components. It primarily consists of loading and thickness noise, and is governed by aerodynamically generated discrete tones at the Blade Passing Frequency (BPF) and its harmonics. One promising control strategy for reducing multirotor noise radiation is ``synchrophasing.'' This approach consists of synchronizing the azimuthal position of the rotor blades across different rotors. This method achieves noise reduction through the destructive interference of the noise generated by different rotors. Because the destructive interference is directional, it is possible to steer this effect toward a specific direction of interest by adjusting the relative phase offsets of each rotor. This method focuses on discrete frequencies, mainly the BPF and its higher harmonics. This dissertation includes a new control method for noise attenuation based on experimental data. It consists of three main steps: \begin{enumerate*}[label=(\roman*)] \item characterization, where each source is identified and acoustically characterized at different speeds and generalized to a single azimuth and elevation angle by using a phase calculation offset based on two parameters (the rotational phase delay and a propagation phase delay) \item optimization, where, based on the identified waveform, the sources are generalized and a set of relative phases generated to provide the best attenuation given a desired target region; \item phase control where the set of relative phases is used as a reference to a regulator controlling multiple rotors\end{enumerate*}. This thesis performs simulations changing vehicle configuration based on characterized data, and noise control authority is presented for four different cases. Acoustic measurements are also taken under diverse conditions, validating the method. Acoustically stationary and non-stationary conditions were evaluated experimentally and by means of simulation. Both conditions included the results for an optimal (most attenuation) case and a pessimal (most amplification) case. The stationary cases kept the phase relation between the rotors constant throughout the test. These include measurements at two different speed set points for fixed observer regions, multiple solutions for a singular region, and measurements for a region all around the vehicle. An evaluation for cases at different elevation angles is included. Finally, an experimental evaluation of the controller performance was conducted, comparing results from the Extended Kalman Filter implementation in hardware and external laser tachometer sensors. The measurement of stationary cases validated the noise prediction and optimization methods presented here. Non-stationary cases changed the observer location or speed set point over time. Measurements indicated that the phase controller was able to track dynamic changes within system parameters and still manage noise levels for both attenuation and amplification. In terms of varying observer location alone, measurements matched the prediction curves, especially when a band pass filter was applied to the raw data with frequency window directed to the first harmonic of the blade passing-frequency. Tests involving varying speed set points demonstrated significant potential of the method, highlighting its capability to track relative phases during the ramp-up of the speed set point. The analysis method was expanded to vary both the observer location as well as to individual motor thrust levels. Finally, single degree of freedom control tests divided rotors into groups to mimic a real single degree of freedom maneuver with the phase controller. Acoustic measurements were taken for roll, pitch and yaw control, demonstrating the feasibility of the method in generating desired flight control moments while directing noise.