See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/381222063

Numerical and Experimental Analysis of Synchronized Propellers for Noise

Mitigation

Conference Paper · May 2024

DOI: 10.2514/6.2024-3237

CITATIONS READS
3 20

8 authors, including:

Burak Buda Turhan Luis Fernando Lopes de Moraes Filho

University of Bristol University of Bristol
17 PUBLICATIONS 54 CITATIONS 2 PUBLICATIONS 3 CITATIONS

SEE PROFILE SEE PROFILE

Shaun Pullin Beckett Yuxiang Zhou

University of Bristol National Institute of Aerospace / NASA Langley Research Center
12 PUBLICATIONS 36 CITATIONS 71 PUBLICATIONS 520 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Burak Buda Turhan on 17 June 2024.

The user has requested enhancement of the downloaded file.

 Numerical and Experimental Analysis of Synchronized
Propellers for Noise Mitigation

Burak Turhan∗ , Luis F. Lopes de M. F.† , Shaun F. Pullin‡ , Beckett Y. Zhou§ , Hasan Kamliya Jawahar¶ , Abhishek
Gautam‖ , Djamel Rezgui∗∗ , and Mahdi Azarpeyvand††
Department of Aerospace Engineering, University of Bristol
Queens Building, University Walk, Bristol, BS8 1TR, United Kingdom

This study investigates the potential of phase synchronization to attenuate the sound power
radiated from multiple Distributed Electrical Propulsion (DEP) configurations under forward
flight conditions. Adopting a combined experimental and numerical approach, the research
aims to analyze how synchronized propellers influence sound radiation, with a particular focus
on the far-field tonal noise and its dependence on the blade phase offset between propellers. A
noise prediction model introduced in this paper is applied to accurately capture sound trends
for various cases of relative phase angles, between adjacent propellers. Acoustic measurements
were conducted in an aeroacoustic wind tunnel using a DEP configuration experimental rig,
which allows adjusting the relative phase angles of two-bladed propellers from Δ𝜓 = 0◦ to
Δ𝜓 = 90◦ . The experiments and numerical simulations were conducted at an advance ratio of
𝐽 = 0.63 and a constant propeller rotation rate of 5000 rpm, where a sound reduction of up to
20 dB was observed at the optimal relative phase angle of Δ𝜓 = 90◦ , compared to the relative
phase of Δ𝜓 = 0◦ . The impacts of relative phases on aerodynamic performance were found to be
negligible, revealing that the noise reduction achieved is primarily due to acoustic interference.
These changes are primarily attributed to far-field acoustic interactions, specifically destructive
interference between the radiated noise from the propellers, which significantly diminishes the
radiated power. This research confirms that phase synchronization is an effective strategy for
noise reduction in DEP systems, leveraging the principle of destructive interference to create a
quieter environment without compromising aerodynamic efficiency.

Nomenclature

𝑐 = wing chord [m]

𝑛 = the rotational speed in revolution per second
𝐵𝑃𝐹 = blade passing frequency [Hz]
𝑉 = free stream velocity [m/s]
𝐷 = propeller diameter [inch]
𝑓 = frequency [Hz]
𝐽 = advance ratio [-]
𝐿 = wing span [m]
𝑟 𝑝𝑚 = revolutions per minute [-]
𝐹𝑥 = resultant load in 𝑥 direction [N]
𝑂 𝐴𝑆𝑃𝐿 = overall sound pressure level [dB]
𝑃 = propeller pitch [inch]
𝑃/𝐷 = propeller pitch to diameter ratio [-]

∗ Research Associate, Department of Aerospace Engineering, University of Bristol, [email protected], AIAA member.
† PhD Candidate, Department of Aerospace Engineering, University of Bristol, [email protected].
‡ PhD Candidate, Department of Aerospace Engineering, University of Bristol, [email protected], AIAA member.
§ Lecturer in Aeroacoustics, Department of Aerospace Engineering, University of Bristol, [email protected]
¶ Research Associate, Department of Aerospace Engineering, University of Bristol, AIAA member (****).
‖ Research Associate, Department of Aerospace Engineering, University of Bristol, AIAA member.
∗∗ Senior Lecturer, Department of Aerospace Engineering, University of Bristol.
†† Professor of Aerodynamics and Aeroacoustics, Department of Aerospace Engineering, University of Bristol.

1
𝑝𝑟 𝑒 𝑓 = reference acoustic pressure [𝜇Pa]
𝑠 = center-to-center-distance [mm]
𝑆𝑃𝐿 = sound pressure level [dB]
𝜓 = relative phase angle [deg]
𝜃 = far-field observer angle [deg]
𝜙 = in-plane observer angle [deg]
r = position vector
𝛼𝐿 = Lamb’s constant
𝜈 = kinematic viscosity
𝑎1 = Squire’s parameter
𝑡 = time
𝛾 = vortex filament strength
𝚪 = vortex particle strength
𝒖 = particle velocity
𝐾 (x, x′ ) = particle kernel
𝑉 = free-stream inflow velocity [m/s]
𝑉𝑥 = volume of particle 𝑥
𝜎 = particle smoothing length
𝜙𝑃𝑃 = power spectral density (dB/Hz)
𝑉 𝐿𝑀 = Vortex Lattice Method
𝑉 𝑃𝑀 = Vortex Particle Method

I. Introduction
The desire for sustainable aviation within the aerospace industry has generated considerable interest in shifting from
fossil fuel-based aircraft propulsion systems to those that are partially or fully electric [1]. A new area that has emerged
in recent years, aiming to leverage advancements in electric propulsion, is urban air mobility (UAM). This field seeks to
provide vehicles capable of transporting people and cargo over short distances, which could be particularly beneficial in
congested urban areas [2]. The UAM vehicles often feature unconventional attributes and designs due to their novel
operating conditions [3]. One such attribute is the distribution of electrically powered propulsive devices across the
vehicle, known as distributed electric propulsion (DEP), which can offer significant aerodynamic advantages such as low
stall speeds and substantial drag reductions [4, 5]. However, one aspect of DEP that is not well understood is how the
proximity of propulsive devices to each other affects the overall aeroacoustic properties of the system. This is particularly
important for UAM vehicles, as noise pollution could impact their societal acceptance [6]. Thus, understanding and
predicting the aeroacoustics of DEP systems is essential for the design process of vehicles utilizing this technology.
In DEP configuration, a series of electric motor-driven propellers along the wing span introduces a unique acoustic
profile distinct from conventional aircraft. A primary factor contributing to this acoustic distinction is the presence
of coherent noise sources. These sources stem from the intricate interactions between the propellers and the wing,
as well as among the propellers themselves [7–9]. Extensive experimental research has been conducted on hover
and forward flight configurations to investigate the impact of various propeller design parameters—such as blade tip
speed, blade pitch, blade radius, chord, twist, and blade count—on aerodynamic and acoustic performance [10–12].
This experimental data was thoroughly analyzed to enhance propeller performance and to identify key aerodynamic
metrics like thrust, drag, efficiency, and the effects of Reynolds number for aerodynamic scaling [13]. Additionally,
comprehensive acoustic and aerodynamic evaluations were carried out to measure aerodynamic loads and far-field
noise radiation, specifically considering edgewise flow conditions [14] and turbulent flow [15]. These aerodynamic
and acoustic effects are largely influenced by the positioning of the propellers relative to the wing or airframe [16–18].
The term ’aerodynamic installation effect’ describes how a propeller interacts with mean-flow gradients around the
wing, leading to additional sources of aerodynamic noise [19, 20]. The acoustic contributions from the force are tonal,
characterized by distinct peaks at the blade passage frequency (BPF) and its harmonics. Impulsive noise may arise from
interactions either between the blades or among neighboring rotors. Additionally, broadband noise results from various
interactions, such as boundary layer turbulence with the trailing edge, turbulence impingement on the leading edge,
and blade vortex interactions. However, even the basic challenge of characterizing noise sources from a single rotor in
forward flight presents a significant difficulty for the scientific community.
In recent years, efforts to reduce the noise levels of multi-rotor platforms have concentrated on active noise reduction.

2
This technique employs phase control to leverage destructive interference between the propellers, treating each propeller
as an independent acoustic source. By adjusting the relative phases of synchronized rotors in DEP configuration, it is
possible to control noise directivity, thus achieving noise reduction in the target direction or region [21, 22]. Using
this method, the noise from one propeller can be adjusted to generate an acoustic wave that effectively reduces the
noise produced by another propeller [21, 23, 24]. Few studies have applied this method to noise reduction in DEP
configurations. Simulation-based analysis has demonstrated the potential to achieve, for a single observation angle, a
significant noise reduction of up to 11 − 30 dB using phase synchronization at an advance ratio 𝐽 = 0.1 [25]. Similarly, a
numerical study focusing on far-field noise prediction revealed the possibility of achieving a substantial 20 dB reduction
in blade passing frequency (BPF) noise at 5000 rpm and 𝐽 = 0.6 using 2-bladed propellers [26]. Recently, Turhan et al.
[21] explored the noise suppression capabilities of DEP systems by assessing their performance under both static thrust
and inflow conditions through electronic propeller synchronization. The study revealed significant reductions in noise
directivity and tonal noise at the first BPF. A relative phase angle of Δ𝜓 = 90◦ achieved the highest noise reduction,
demonstrating an 8 dB decrease at the first BPF and a 2 dB decrease in the overall sound pressure level (OASPL) under
static thrust conditions (𝐽 = 0). For inflow conditions (𝐽 > 0), this phase angle resulted in a dramatic reduction of
approximately 24 dB at the first BPF and 6 dB in overall sound pressure level (OASPL), compared to Δ𝜓 = 0◦ .
Phase control, or phase synchronization, involves ensuring that propellers rotate synchronously at equivalent rates.
This technique focuses on adjusting the relative azimuthal blade positions, or phases, to minimize the noise associated
with blade passage frequencies. This study aims to determine numerically whether controlling the relative phase of
matched rotor pairs can achieve a net acoustic benefit across all directions. Our efforts are focused exclusively on
understanding the dynamics under inflow conditions (𝐽 > 0). This study encompasses both experimental and numerical
assessments of two adjacent, 2-bladed propellers. The paper begins with a detailed description of the experimental
setup and the computational model employed. Results are then presented, starting with measured spectra and source
directivity. Numerical results are discussed to justify the far-field approximation used in calculating radiated sound power
from pressure measurements. Predicted and measured sound power levels are compared, and a physical explanation is
proposed.

II. Methodology
In this section, a comprehensive description of the methodologies employed in the study is provided, encompassing
both experimental and numerical approaches. The experimental procedure involves systematic testing with 2-bladed
propellers using electronically phase-locked propellers. On the numerical side, a sophisticated simulation tool is utilized
to model the aerodynamic and acoustic phenomena observed during the experiments. Together, these methodologies form
the backbone of the research, aiming to produce validated results that advance the understanding of propeller-induced
noise and its mitigation.

A. Experimental Rig Setup and Test Matrix

The experiments were carried out at the University of Bristol’s aeroacoustic wind tunnel facility, which is a
temperature-controlled, closed-circuit configuration for a free-stream range from 15 m/s to 40 m/s, with a turbulent
intensity as low as ≈ 0.2%. The exit nozzle measures 0.5 m in width and 0.775 m in height, with a contraction ratio
of 8.4. The anechoic room’s exposed surfaces, including the anechoic chamber and contraction nozzle, are either
acoustically lined or covered in acoustic foam wedges to completely absorb sound reflections. According to the ISO
3745 standardized testing procedure for both pure tone and broadband testing, the anechoic chamber allows for anechoic
measurements down to 160Hz [27].
The experimental setup, as illustrated in Fig. ??, consists of two propulsion units with propellers attached to a
wing, representative of a DEP configuration. The tests employed 2-bladed propellers, each with a diameter of 𝐷 = 9”
and a pitch-to-diameter ratio of 𝑃/𝐷 = 1 in a co-rotating configuration. This study solely focuses on the co-rotating
configuration. Co-rotating propellers appear to offer more advantages in terms of noise reduction than counter-rotating
propellers when using phase synchronization [22, 28]. Therefore, the counter-rotating configuration is not considered.
These propellers were custom designed and manufactured by Mejzlik Propellers (https://www.mejzlik.eu/) using carbon
fiber/epoxy-based materials, with a Clark-Y airfoil section. Details about the blade chord and the distribution of pitch
angles along the blade span can be found in Fig. 2. The wing was designed using a NACA0018 profile, with a chord
of 𝑐 = 0.3 m and a span of 𝐿 = 0.94 m and manufactured using 6000 series aluminum alloy. The propellers were
positioned 150 mm from the wing’s trailing edge to the propeller’s leading edge, and roughly 0.5 m from the nozzle exit,
as is normal with the tunnel free-stream flow.

3
 In-plane microphones
(a) (simulation)

1.75m

Far-field
microphone array
Flow

(b) Propeller 1 Propeller 2

Dy

Reference
position
s

Fig. 1 (a) Schematic representation of the experimental setup, (b) relative phase synchronization for the
two-propeller configuration.

Fig. 2 Distribution of the blade chord length (𝑐/𝑅) and the pitch angle for the 2-bladed 9” × 9” propeller used
in this study.

4
 A schematic of the far-field arrays used in the present study along with the experimental setup is demonstrated in
Fig. 1 (a). The far-field array was placed with the 𝜃 = 90◦ microphone position at a distance of 1.75 m above the center
of the wing and symmetrically separated between the two propellers. The overhead array consists of 18 G.R.A.S 40PL
microphones (1/4 inch diameter) each spaced at 5◦ intervals between polar angles of 55◦ and 130◦ . The microphones
have a frequency range of 𝑓 = 10 Hz to 20 kHz and dynamic ranges of 142 dB with an accuracy of ±1 dB. The data
were acquired at a sampling frequency of 216 Hz and a measurement time of 𝑡 = 40 s. The microphones were calibrated
using a G.R.A.S 42AA pistonphone calibrator before and after the experiments. The power spectra are constructed
using the MATLAB Pwelch method [29], and the results are presented with a frequency resolution of 1 Hz.
Each propeller in the setup is powered by a 540kV AT4125 T-motor with a maximum continuous output of 2.2 kW.
To track the phase angle of the propeller blades with high precision, a 12-bit RLS® RE36IC incremental output encoder
is employed with an accuracy of ±0.3◦ . The propellers’ relative phase angles are managed through a specialized system
that sends two distinct signals for different purposes. The first signal is used by the motor control software to regulate the
propellers’ relative phase angles. The second signal is sent to a Tektronix TBS1102C oscilloscope, allowing for real-time
monitoring of any changes in phase angle during tests. Additionally, Tektronix electric voltage probes are connected to
capture the position signal, which is generated by the controller after processing pulses from the position encoder. At
an operating speed of 5000 rpm, the setup maintains a positional error consistently below 1◦ , and the peak-to-peak
speed error is kept under 0.05% (2.5 rpm) [21]. Note that this drivetrain setup is used for practical purposes because
electronically controlled phase matching is expected to be used in practice.
Figure 1 (b) presents the definition of the relative phase angle (Δ𝜓) for the two propeller systems and the other
geometric parameters of the experimental setup. In this setup, the spacing between the propellers, referred to as the
center-to-center separation distance, was set to a ratio of 𝑠/𝐷 = 1.05. The relative phase angle is defined by the
angular difference between the blades of the two propellers. To establish a baseline, the propellers were initially aligned
horizontally along the wing span, setting the initial relative phase difference at Δ𝜓 = 0◦ . Propeller 1 was designated
as the master propeller, and Propeller 2 was termed the slave propeller. The blade position on Propeller 2 was then
adjusted to achieve various relative phase angles, while Propeller 1 was at its original position. This study investigated
seven different relative phase angles (Δ𝜓) ranging from 0◦ to 90◦ in increments of 15◦ . The aeroacoustic measurements
were conducted at a wind tunnel velocity of 12 m/s, which corresponds to an advance ratio of 𝐽 = 0.63 at the propellers
rotating at a speed of 5000 rpm. This advance ratio was selected as one of the advance ratios for the experiments due to
its optimal operational efficiency. The advance ratio is a non-dimensional term defined as:
𝑉
𝐽= (1)
𝑛𝐷
where 𝑉 is the free-stream inflow velocity in m/s, 𝑛 the rotational speed in revolution per second, and 𝐷 is the propeller
diameter in meters.

B. Mid-Fidelity Aerodynamic and Aeroacoustic Simulation

The aerodynamic and aeroacoustic coupled characteristics between two phase-synchronized propellers were modeled
using a mid-fidelity computational aerodynamic and aeroacoustic tool, as depicted in the configuration shown in
Fig. 1. Additionally, in-plane microphone arrays were placed 1.75 meters away from the center-to-center location of
the separation distance to facilitate numerical analysis, as shown in Fig. 1(a). This tool employs unsteady, potential
flow-based methods to simulate aerodynamic surfaces using the Vortex Lattice Method (VLM). The VLM models thin
surfaces with a grid of quadrilateral vortex ring elements composed of constant-strength vortex filaments. The velocity
induced by each vortex filament with strength 𝛾 and end position vectors at r1 and r2 can be determined at an evaluation
point r0 by using the Biot-Savart law with a finite vortex core size.

𝛾 (∥ 𝒓 1 ∥ + ∥ 𝒓 2 ∥) (∥ 𝒓 1 ∥ ∥ 𝒓 2 ∥ − 𝒓 1 · 𝒓 2 )
u(x, 𝑡) = , (2)
4𝜋 ∥ 𝒓 1 ∥ ∥ 𝒓 2 ∥ (∥ 𝒓 1 × 𝒓 2 ∥ 2 + 𝑟 𝑐2 ∥ 𝒓 2 − 𝒓 1 ∥ 2 )
√
where 𝑟 𝑐 = 4𝛼 𝐿 𝜈𝛿𝑡 and 𝛿 = 1 + 𝑎 1 𝛾𝜈 Each element has a collocation point at which a Neumann boundary condition
enforces no flow velocity normal to the surface. This condition is used to determine the strengths of each panel by
assembling and solving a linear system of equations that relates the strength of each panel to the flow velocity induced at
each collocation point.
Unlike traditional unsteady VLM and panel methods, the wake from an aerodynamic surface is not formed from a
deformable sheet of vortex ring elements. Instead, it is modeled using vortex particles adhering to the Vortex Particle

5
Method (VPM). This method provides a Lagrangian description of wake vorticity, defined as discrete, moving packets
of fluid (particles). The main advantage of replacing a paneled wake with a particle-based wake is that it eliminates
issues surrounding mesh entanglement and panel distortion. These issues can become prevalent with wake-wake and
wake-surface interactions, which may be common in multi-surface problems. The VPM is derived from a discrete form
of the Vorticity Transport Equation, which, in turn, can be derived from the Navier-Stokes momentum equation by
assuming the flow is incompressible. A further assumption that the velocity field induced by the vortex particles is
a solenoidal (divergence-free) field allows the solution for the said field to be determined through the use of Green’s
function. It also permits the velocity field induced by the surface VLM elements, which is irrotational (curl-free), to
be combined by superposition with the particle-induced velocity through Helmholtz’s decomposition. The scheme
that defines the VPM used for the numerical simulation in this work will be presented below. For more details and
derivations, see [30–32].

d d
𝚪 𝑝 (𝑡) = (𝚪 𝑝 (𝑡) · ∇)u(x 𝑝 (𝑡), 𝑡) + 𝚪visc , (3)
d𝑡 d𝑡 𝑝
d ∑︁
x 𝑝 (𝑡) = u(x 𝑝 (𝑡), 𝑡) = 𝐾 (x 𝑝 (𝑡), x𝑞 (𝑡)) × 𝚪 𝑝 (𝑡), (4)
d𝑡 𝑞
d visc 105 ∑︁ 𝑉 𝑝 𝚪𝑞 − 𝑉𝑞 𝚪 𝑝
𝚪𝑝 = 5  92 . (5)
d𝑡 4𝜋𝜎 𝑞  x −x 
∥ 𝑝 𝑞∥ 2

𝜎 𝜎 +1

For the method used in this work the particle kernel, 𝐾 (x, x′ ), is a Winkelmanns and Leonard [31] kernel. The viscous
component of the VPM scheme is evaluated using a Particle Strength Exchange scheme. A new row of particles is
initialized at the trailing edge of each aerodynamic surface every time step, with their strengths determined by the
strengths of the vortex rings at the trailing edge in the previous time step. Once an aerodynamic solution has been
determined, the acoustic propagation from surfaces in the fluid domain is established using Farassat’s Formulation 1A
[33] of the Ffowcs-Williams and Hawkings equation. Both the monopole and dipole sources are integrated from the
pressure distribution on the aerodynamic surfaces, while quadrupole contributions from the volume are neglected. All
of the computational methods presented above have been implemented in a single framework that utilizes GPU devices
for efficient, large-scale vectorization and parallelization, providing a fast time-to-solution for each problem.
For this study, the wing was modeled with a uniform distribution of 80 span-wise panels and a cosine distribution of
16 chord-wise panels for each of the upper and lower surfaces, resulting in a model with 2560 panels. Each propeller
was modeled with two blades, each comprising 41 span-wise panels and 24 chord-wise panels, which amounts to a
model with 984 panels per blade or 1968 panels per propeller. Thus, the complete Vortex Lattice Method (VLM) model
comprised 6496 panels. In the time domain, all simulations were run for a total of 15 revolutions of the propellers,
resulting in a total physical time of 0.18 s. The timestep used for the simulations corresponded to a 2◦ rotation of the
propeller, leading to 2700 time steps each with a duration of 6.67 × 10−5 s. Figure 3 displays the time history of the
load (𝐹𝑥 ), parallel to the propellers’ axes, computed on the model throughout the simulation. Due to the dominance of
propeller thrust in the set, the (𝐹𝑥 ) load component was utilized to determine when the flow is considered stabilized.
Following this criterion, aerodynamic and noise measurements were taken during the last 10 revolutions, specifically
from complete revolution number 6 to complete revolution number 15, corresponding to the time range from 0.06 s to
0.18 s.

III. Results and Discussion

Tests were conducted for twin-propeller configurations in DEP using a phase-locked control system. First, the
influence of phase synchronization on the noise directivity and frequency characteristics was analyzed using experimental
data. Based on this, the noise reduction achieved through phase control verifies the effectiveness of the numerical
approach.

A. Experimental Results
This section presents the experimental findings, beginning with the far-field acoustic results. The analysis focuses
on comparing the spectral and directivity characteristics at the first BPF of the phase-locked propellers across various

6
Fig. 3 Time history of component load parallel to propellers axes for the complete setup (Δ𝜓 = 0◦ ). The vertical
dashed line indicates the time from which aerodynamics and noise quantities are calculated.

relative phase angles, with a free-stream velocity of 12 m/s, corresponding to the advance ratio of 𝐽 = 0.63. Following
this analysis, we explore the directivity of the overall sound pressure level (OASPL) to evaluate the effectiveness of the
phase synchronization method in mitigating noise levels.

Fig. 4 Far-Field SPL for two 2-bladed propellers various phase angles at an advance ratio 𝐽 = 0.63 for observer
angle 𝜃 = 90◦ (a) noise spectra and (b) the first BPF tones.

Figure 4 displays the sound pressure level (SPL) across a wide frequency range observed at an angle of 𝜃 = 90◦ ,
illustrating the noise spectrum for seven phase angles (Δ𝜓). The SPL was calculated using the following equation:
!
𝜙𝑝𝑝
𝑆𝑃𝐿 = 10 · log10 2 , (6)
𝑝𝑟 𝑒 𝑓
where 𝜙 𝑝 𝑝 represents the power spectral density of the measured acoustic pressure, and 𝑝 𝑟 𝑒 𝑓 is the reference acoustic
pressure (equal to 20 𝜇Pa).
Figures 4 (a) and (b) show the sound pressure level spectra measured at the observation point (𝜃 = 90◦ ). The
motor mechanical noise, illustrated by the dashed light grey line, consists of distinct tones layered over a general
broadband noise level. It is evident that the aerodynamic noise generated by the propeller surpasses the mechanical
noise from the motor by a margin of over 15 dB. This suggests that the motor’s contribution to the overall noise profile
is minimal, confirming that the aerodynamic noise from the propeller is the dominant factor in the acoustic readings

7
captured by the far-field microphone array. The tonal noise is identified as the first BPF, at 𝑓 = 166.6 Hz, along with its
harmonics, which occur periodically with each propeller rotation. Compared to Δ𝜓 = 0◦ , setting the relative phase
to Δ𝜓 = 90◦ resulted in a significant reduction in noise levels at most harmonic frequencies, especially at the BPF,
except for a very small number of harmonic frequencies where the noise level shows a different trend. The results
indicate that adjusting the blade’s relative phase angle (Δ𝜓) leads to negligible fluctuations in the broadband noise
levels. However, the results demonstrate that altering the blade’s relative phase angle from Δ𝜓 = 0◦ to 90◦ can result in
significant changes to the magnitude of the first BPF. This is consistent with previous investigations on propeller phase
synchronization [21, 24, 34, 35].
Fig. 4(b) illustrates the noise levels of the first BPF as observed from an observer angle of 𝜃 = 90◦ , comparing seven
different relative phase angles at an advance ratio of 𝐽 = 0.63. Generally, it was found that a relative phase angle of
Δ𝜓 = 90◦ leads to a marked decrease in the BPF noise level compared to a relative phase angle of Δ𝜓 = 0◦ . Specifically,
the sound pressure level for the Δ𝜓 = 90◦ phase angle is about 18 dB lower than that for the phase angle of Δ𝜓 = 0◦ .
Evaluating the results across various relative phase angles Δ𝜓, it becomes apparent that there is no substantial noise
reduction when the relative phase difference is less than Δ𝜓 < 30◦ , in relation to the baseline position of Δ𝜓 = 0◦ .
However, when the relative phase difference exceeds Δ𝜓 > 45◦ , there is a noticeable step-wise reduction in the first BPF
with each increase in phase angle.

Fig. 5 SPL directivity pattern at the first BPF of two 2-bladed propellers with seven relative phase angles, for
an advance ratio of 𝐽 = 0.63.

The results in Fig. 5 highlight the substantial impact that the relative phase angle of the blades has on the amplitude
of the first BPF, as well as the influence of the observer’s position (𝜃). A marked decrease in the first BPF noise is noted
when the relative phase angle is altered, affecting the entire directivity range. The most significant noise reduction
consistently occurs at a relative phase angle of Δ𝜓 = 90◦ , regardless of the observer’s angle. Analysis of the results for
varying relative phase angles (Δ𝜓) indicates that when the relative phase difference is less than Δ𝜓 < 30◦ , it does not
yield a significant reduction in noise compared to the baseline relative phase angle of Δ𝜓 = 0◦ across all observer angles.
However, for phase differences greater than Δ𝜓 > 45◦ , there is a discernible step-by-step decline in the SPL with each
increase in relative phase angle. Additionally, the results reveal that at a relative phase of Δ𝜓 = 75◦ , the noise reduction
is predominantly at downstream angles. In contrast, a relative phase angle of Δ𝜓 = 90◦ is associated with significant
noise reduction across the spectrum of observer angles. In conditions of forward flight at a constant speed, the phase
synchronization method can be employed to alter the noise directivity emitted by the DEP configuration. Adjusting the
angular positions of the blades demonstrates significant potential as a noise-canceling technique through destructive
wave interference.
Figure 6 presents the directivity of the overall sound pressure level (OASPL) over 55◦ ≤ 𝜃 ≤ 130◦ for seven relative
phase angles. The OASPL analysis assesses the amount of noise generated and examines how the interaction of the
propellers affects directivity. The OASPL is calculated by integrating the SPL across a specific frequency range (from
100 Hz to 1000 Hz), as shown below:

8
 Í !
𝜙 𝑃𝑃 ( 𝑓 ) · Δ 𝑓
OASPL = 10 · log10 . (7)
𝑝 2ref
In general, a considerable reduction in OASPL is observed with the increase of the relative phase angle (Δ𝜓),
affecting all directivity angles. Altering the relative phase angle between the propellers to Δ𝜓 = 90◦ leads to a drop in
noise levels, yet it does not appear to significantly alter the directivity pattern of the OASPL. Similar to previous results
of the first BPF, significant noise reduction becomes apparent only when the relative phase angle exceeds Δ𝜓 > 30◦ .
The noise reduction increases, peaking at a maximum attenuation of up to 6 dB when the relative phase angle is set to
Δ𝜓 = 90◦ .

70 =0° =45 ° =90 °

° °
=15 =60

60 =30 ° =75 °

50

40

30
40 60 80 100 120 140

Fig. 6 Overall sound pressure level directivity of two 2-bladed propellers with seven relative phase angles for an
advance ratio of 𝐽 = 0.63.

B. Numerical Results

1. Isolated Propeller Benchmark

Initial validation results for the Panel/VPM method were obtained by simulating a benchmark propeller as described
by Casalino et al. [36]. The propeller, derived from a 2-bladed APC-96 model with a NACA 4412 profile and a rescaled
diameter of 0.3 m, had its geometric parameters provided. Thrust measurements were obtained under uniform inflow
conditions varying advance ratios from 𝑉 = 0.0 to 20.0 m/s (𝐽 = 0.0 to 0.8 at 5000 rpm). This case was modeled using
vortex lattice elements and vortex particles, covering all experimentally measured advance ratios. The simulation was
performed over 10 revolutions, with the global loads reaching a steady state.
Figure 7 compares the results of the panel/VPM model with the experimentally measured thrust and torque.
Additionally, it includes thrust predictions using a finite volume method with a RANS-based turbulence model, executed
using the SU2 solver. The thrust is accurately predicted by the mid-fidelity method and provides a level of accuracy
similar to that of the experiment when compared to the more computationally intensive finite-volume method.
The aeroacoustic performance of this implementation has been validated by correlating numerical simulations with
analytical results for both an acoustic monopole and dipole. The results of this comparison are presented in Fig. 8,
which shows that the numerical prediction exhibits excellent agreement with the analytical results. This aeroacoustic
implementation has also been used in previous studies [37] that have performed propeller simulations and have provided
very good agreement with experimental acoustic observations.

2. Noise reduction performance of synchronized twin propellers

In this section, the numerical results for twin synchronized propellers with a wing are elucidated and visualized
in Figure 9. For all simulations, the free-stream velocity was set to 12 m/s, and the propellers’ rotation to 5000 rpm,
corresponding to an advance ratio of 𝐽 = 0.63, matching the experimental parameters. The simulated relative phase
angles (Δ𝜓), ranging from 0◦ to 90◦ in 15◦ increments, those used in the experiments. Figure 9 provides a perspective
view of instantaneous flow structures, characterized by vortical patterns emanating from the propeller blades’ tips.
The wake, generated using the Vortex Panel Method for in-phase propellers (Δ𝜓 = 0◦ ), features vortical structures
following a helical path from the blade tip, interacting with the wing, and propagating downstream. The simulation also

9
 10
Experiment
8 VLM
RANS (SU2)

Thrust [N] 6

-2
0 5 10 15 20
Velocity [m/s]

Fig. 7 Comparison of experimentally measured loads for a benchmark propeller to various numerical simulation
techniques including the VLM/VPM method.

Fig. 8 Analytical verification of aeroacoustic solver using a monopole and dipole source.

10
reveals how the interaction with the wing deforms the vortex structures, an effect that becomes more pronounced further
downstream.

Fig. 9 Overview of the DEP configuration with VLM panel modeling and the particle wake from VPM
simulation.

Figure 10 displays the time-averaged pressure distribution on a blade for each propeller’s angular position, calculated
for a range of relative phase angles on the suction surface of the left propeller. For better visualization of the aerodynamic
effects caused by the interaction between the propellers, the time-averaged pressure was subtracted by the mean value
radius-wise, as shown in the bottom-right plot of Fig. 10. The resulting plots underscore that, while the methodology
effectively captures the aerodynamic interactions—evident from the reduced low-pressure area near the propeller
interface with increasing phase angle—the magnitude of pressure fluctuations (ranging from −37.5 Pa to 30.0 Pa) is
minor relative to the mean pressure level (approximately 100 kPa). This implies that noise behaviors are likely more
affected by aeroacoustic than by aerodynamic pressure effects. These findings are consistent with existing literature.
Schiller et al. [24] indicate some variation in measured load values, especially for the counter-rotating case. However,
these variations suggest that the impact of the relative phase angle on aerodynamic performance is generally negligible.
This is further supported by Zarri et al. [35], who found that the influence of the relative phase angle on aerodynamic
performance results in a minimal change in the thrust coefficient, less than 0.75%. The main objective of this study
is to attribute the observed noise reduction primarily to the acoustic interference between the two propellers, rather
than to aerodynamic loading effects. This inference is supported by the premise that acoustic interference, particularly
influenced by the mismatching of tip vortices, plays a pivotal role in noise modulation under specified experimental
conditions.
Figure 11 compares the first BPF SPL directivity patterns for a selected range of relative phase angles derived
from numerical simulations with those obtained from experiments, as shown in Figure 5. The numerical simulations
employed the same observer positions as the experiments, following the schematics provided in Figure 1(a). While the
primary goal of the experiment was to collect data that could be used to assess the accuracy of the numerical model, the
experimental data can also directly quantify the change in radiated sound power when the microphones are positioned in
the far field. In general, there appears to be more scatter between the measurements and numerical results for relative
phase angles after Δ𝜓 ≥ 60◦ . However, results for phase angles up to Δ𝜓 < 60◦ tend to be consistent. Additionally,
both experimental and numerical results demonstrate a clear correlation between an increased relative phase angle
and noise reduction in the far-field. The most notable acoustic attenuation was observed at a relative phase angle
of Δ𝜓 = 90◦ , where reductions in the first BPF amplitudes were significantly higher compared to those at Δ𝜓 = 0◦ ,
across all directivity angles. Generally, both the shape and magnitude of sound directivity exhibit strong similarities

11
Fig. 10 Time-averaged pressure distribution on the disc representing the suction surface of the left propeller for
various relative phase angles.

between experimental and numerical results when considering upstream directivity angles 𝜃 < 90◦ . In contrast, for
downstream observers 𝜃 ≥ 90◦ , numerical methods consistently underestimate the noise level, with the discrepancies
widening as the distance of the observer increases. Furthermore, there is a noted inconsistency in the relative phase
Δ𝜓 = 75◦ throughout the directivity angle spectrum, with numerical predictions being approximately 4 dB lower than
those observed experimentally.

Fig. 11 SPL directivity pattern for the first BPF octave band, obtained from numerical simulation (lines),
compared with the pattern at the first BPF from experimental results (markers) for seven relative phase angles at
the advance ratio of 𝐽 = 0.63.

The primary goal of this study is to explore the impact of the propellers’ relative phase angle on noise reduction, both

12
numerically and experimentally. This is achieved by comparing the noise levels at the first BPF using each relative phase
angle and a reference relative phase of Δ𝜓 = 0◦ . The term ’ΔSPL’ denotes the difference in SPL between each relative
phase angle and Δ𝜓 = 0◦ , serving as a crucial measure of the effectiveness of noise reduction. Figure 12 displays a
comparative analysis of the ΔSPL at the first BPF across all relative phase angles, incorporating both experimental and
numerical results. The numerical model effectively replicates the noise reduction up to a relative phase of Δ𝜓 < 60◦ .
However, for relative phases Δ𝜓 ≥ 60◦ , it predicts greater noise reduction than observed experimentally. Specifically, at
a relative phase angle of Δ𝜓 = 75◦ , a maximum discrepancy of 13 dB is noted in the upstream region, tapering off to
4.5 dB at the propeller plane and to 2 dB downstream. At Δ𝜓 = 90◦ , the disparity between the numerical predictions
and experimental measurements narrows to approximately 4 dB for upstream observer angles and reduces further to
2 dB for downstream angles.

Fig. 12 Comparison of ΔSPL variation between each relative phase angle and the propeller reference configura-
tion (Δ𝜓 = 0◦ ) across directivity angles for both experimental (markers) and numerical (lines) simulation data.

Figure 13 displays the SPL directivity in-plane azimuthal levels at the first BPF, as calculated by a numerical model.
This plot illustrates the isolated effects on SPL for the right and left propellers, the combined effects of both propellers
and the results for the DEP configuration. The analysis demonstrates how the relative phase angle affects sound
levels when both propellers are operating. It also shows that the sound levels for the entire model are predominantly
influenced by the effects of the propellers, with the wing’s contribution being so negligible that it does not appear
on the presented scale. Initially, Fig. 13 suggests that the numerical model can accurately predict the noise of an
isolated rotor. Regardless of the relative phase angle, both the right and left propellers alone exhibit the same noise
level, as expected. From Δ𝜓 = 0◦ , it is evident that the combination of two propellers achieves a maximum increase of
approximately (6 dB) compared to a single isolated propeller. This increase is anticipated due to the overlay of two
in-phase signals. Additionally, when the rotors are set with relative phase offsets of 0◦ and 15◦ , the directivity becomes
nearly axisymmetric. Azimuthal asymmetry is observed at relative phases of Δ𝜓 = 30◦ and Δ𝜓 = 45◦ . Despite this
asymmetry, the radiation pattern still conforms to a classic monopole configuration. In contrast, when the relative
phase angle reaches Δ𝜓 = 60◦ , there is a significant reduction in sound levels across the complete setup, affecting a
broad range of directivities from 90◦ to 225◦ . When the relative phase angle reaches Δ𝜓 = 75◦ , a cardioid pattern
emerges, and the noise reduction spans a broad range of directivities from 45◦ to 275◦ . Finally, when the relative phase
offset reaches 90◦ , it produces a figure-8 pattern with two lobes separated by deep at approximately 60◦ and 250◦ ,
characteristic of a classic dipole radiation pattern. Beyond the alteration in shape, there is also a notable reduction in
overall amplitude, even in directions of peak radiation. Overall, the numerical model effectively captures the essential
source characteristics, including variations in both the shape and amplitude of the radiation patterns.
Given that the reduction or amplification of a signal results primarily from the destructive or constructive interactions
among overlaid signals, Figures 14 and 15 illustrate these interactions for a study involving synchronized twin propellers.

13
Fig. 13 The directivity of the first BPF is computed on the plane of the propellers for seven relative phase angles.
Each plot shows contributions from isolated and combined propellers.

14
Figure 14 displays pressure fluctuation contours at an arbitrary instant in time for a region of the propeller plane and four
relative phase angles Δ𝜓 = 0◦ , 45◦ , 75◦ , and 90◦ . The coordinates of these contours have been dimensionless relative to
the propellers’ diameter, and vertical dashed lines denote the rotors’ centerline, which is perpendicular to their axes.
For relative phase angles Δ𝜓 = 0◦ and 45◦ , the wave patterns are distinctly visible, allowing for the identification of
peaks, valleys, and nodes. These features’ clarity and amplitude suggest a constructive interaction, especially prominent
at Δ𝜓 = 0◦ . However, at Δ𝜓 = 45◦ , the wave nodes appear more dispersed, indicating a decrease in constructive
interaction. For Δ𝜓 = 75◦ , although peaks, valleys, and nodes can be identified, the nodes are so sparse that they nearly
flatten the entire region. Even so, it is evident that in the depicted waves, the right region exhibits more pronounced
peaks and valleys than the left. This observation aligns with the findings in Figure 13, which shows a drop in SPL in the
left region. Finally, at Δ𝜓 = 90◦ , the nodes are indistinct, reflecting the extensive disruption of wave structures that
results in a significant reduction of SPL.

Fig. 14 Instant pressure fluctuation in the plane of the propellers for relative phase angles Δ𝜓 = 0◦ , 45◦ , 75◦
and 90◦ .

To illustrate the noise reduction benefits of phase control, Figure 15 presents the SPL, calculated for the same region
using the time history of pressure fluctuations over three complete revolutions of the propellers. Variations in pressure
fluctuations between the relative phases of Δ𝜓 = 0◦ and Δ𝜓 = 90◦ may explain the observed reductions in SPL. One
explanation for the change in power at Δ𝜓 = 90◦ is that the far-field acoustic interaction alters the radiation efficiency,
and thus the net radiated power of the twin-propeller system. This accounts for the dipole shape of the directivity,
which consists of two closely spaced monopoles with the same source strength but opposite phases and radiates less
efficiently than a single monopole. These results indicate that phase synchronization is a viable method for achieving
a net reduction in acoustic radiation at the first BPF from two-propeller configurations. As demonstrated above, the
effectiveness of noise reduction at the first BPF depends on the relative phase angle of the blades, both experimentally
and numerically.

IV. Conclusion
This study aimed to assess whether phase synchronization could reduce the sound power radiated by the DEP
configuration. Incorporating both experimental and numerical approaches, the results confirmed that the model
effectively captures the sound radiation from two synchronized propellers. Subsequently, the model facilitated the
exploration of the underlying mechanisms and evaluation of the advantages of employing phase synchronization in pairs
of matched rotors.
The results of this study show that the far-field tonal noise is strongly influenced by the phase offset between the

15
 Fig. 15 Contours of SPL in the plane of the propellers for relative phase angles Δ𝜓 = 0◦ , 45◦ , 75◦ and 90◦ .

propellers. The experiment and numerical study were conducted at an advance ratio of 𝐽 = 0.63 with a constant
propeller rotation rate of 5000 rpm. Both measurements and numerical predictions demonstrate that the radiated sound
power at the blade passage frequency can be reduced by synchronizing and controlling the relative phase of two rotors.
A reduction of 19 dB was achieved experimentally, while numerical analysis showed a 20 dB reduction. The optimal
relative phase for two-bladed propellers is Δ𝜓 = 90◦ , which was confirmed both experimentally and numerically.
The pressure fluctuations begin to change at a relative phase of Δ𝜓 = 45◦ and reach minimum values at Δ𝜓 = 90◦ .
Consequently, noise reduction does not exceed 5 dB until Δ𝜓 = 45◦ . The reduction in power at Δ𝜓 = 90◦ is found to be
due to far-field acoustic interactions that decrease radiation efficiency, thereby reducing the net radiated power of the
twin-propeller system and resulting in a dipole directivity pattern. This pattern is less efficient than radiation from a
single monopole.
The impact of the relative phase angle on aerodynamic performance is negligible, confirming that the observed
acoustic effects are purely due to acoustic interference. The findings demonstrate that phase synchronization is an
effective noise reduction technique in DEP systems, utilizing the principle of destructive interference within the coherent
acoustic source field between the propellers.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have
appeared to influence the work reported in this paper.

Data Availability
Data will be made available on request.

Acknowledgement
The authors gratefully acknowledge financial support from the Horizon 2020 research and innovation program
(grant agreement number 882842) for the SilentProp project.

16
 Bibliography
[1] Sahoo, S., Zhao, X., and Kyprianidis, K., “A Review of Concepts, Benefits, and Challenges for Future Electrical Propulsion-Based
Aircraft,” Aerospace, Vol. 7, No. 4, 2020.

[2] Kim, H. D., Perry, A. T., and Ansell, P. J., “A review of distributed electric propulsion concepts for air vehicle technology,”
2018 AIAA/IEEE Electric Aircraft Technologies Symposium (EATS), IEEE, Cincinnati, OH, US, 2018, pp. 1–21.

[3] Straubinger, A., Rothfeld, R., Shamiyeh, M., Büchter, K.-D., Kaiser, J., and Plötner, K. O., “An overview of current research and
developments in urban air mobility–Setting the scene for UAM introduction,” Journal of Air Transport Management, Vol. 87,
2020, p. 101852.

[4] Gohardani, A. S., Doulgeris, G., and Singh, R., “Challenges of future aircraft propulsion: A review of distributed propulsion
technology and its potential application for the all electric commercial aircraft,” Progress in Aerospace Sciences, Vol. 47, No. 5,
2011, pp. 369–391.

[5] Stoll, A. M., Bevirt, J., Moore, M. D., Fredericks, W. J., and Borer, N. K., “Drag reduction through distributed electric
propulsion,” 14th AIAA aviation technology, integration, and operations conference, 2014, p. 2851.

[6] Al Haddad, C., Chaniotakis, E., Straubinger, A., Plötner, K., and Antoniou, C., “Factors affecting the adoption and use of urban
air mobility,” Transportation research part A: policy and practice, Vol. 132, 2020, pp. 696–712.

[7] Turhan, B., Kamliya Jawahar, H., Bowen, L., Rezgui, D., and Azarpeyvand, M., “Aeroacoustic characteristics of distributed
electric propulsion system in forward flight,” AIAA AVIATION 2023 Forum, San Diego, CA, US, 2023, p. 4490.

[8] Bowen, L., Turhan, B., Kamliya Jawahar, H., Rezgui, D., and Azarpeyvand, M., “Aeroacoustic characteristics of distributed
electric propulsion configuration with turbulent flows,” AIAA AVIATION 2023 Forum, San Diego, CA, US, 2023, p. 4294.

[9] Zhou, W., Ning, Z., Li, H., and Hu, H., “An experimental investigation on rotor-to-rotor interactions of small UAV propellers,”
35th AIAA applied aerodynamics conference, 2017, p. 3744.

[10] Baskaran, K., Jamaluddin, N. S., Celik, A., Rezgui, D., and Azarpeyvand, M., “Effects of number of blades on propeller noise,”
Journal of Sound and Vibration, Vol. 572, 2024, p. 118176.

[11] Baskaran, K., Jamaluddin, N. S., Celik, A., Rezgui, D., and Azarpeyvand, M., “Aerodynamic and aeroacoustic characteristics of
propellers with different blade numbers,” 28th AIAA/CEAS Aeroacoustics 2022 Conference, 2022, p. 3108.

[12] Baskaran, K., Celik, A., Jamaluddin, N. S., Rezgui, D., and Azarpeyvand, M., “Aerodynamic and aeroacoustic characteristics of
different pitch propellers,” 28th AIAA/CEAS Aeroacoustics 2022 Conference, 2022, p. 3107.

[13] Deters, R. W., Ananda Krishnan, G. K., and Selig, M. S., “Reynolds number effects on the performance of small-scale
propellers,” 32nd AIAA applied aerodynamics conference, 2014, p. 2151.

[14] Jamaluddin, N. S., Celik, A., Baskaran, K., Rezgui, D., and Azarpeyvand, M., “Aerodynamic noise analysis of tilting rotor in
edgewise flow conditions,” Journal of Sound and Vibration, 2024, p. 118423.

[15] Jamaluddin, N. S., Celik, A., Baskaran, K., Rezgui, D., and Azarpeyvand, M., “Experimental Analysis of Rotor Blade Noise in
Edgewise Turbulence,” Aerospace, Vol. 10, No. 6, 2023.

[16] Jamaluddin, N. S., Celik, A., Baskaran, K., Rezgui, D., and Azarpeyvand, M., “Experimental characterisation of small-scaled
propeller-wing interaction noise,” 28th AIAA/CEAS Aeroacoustics 2022 Conference, 2022, p. 2973.

[17] Paruchuri, C. C., Akiwate, D. C., Palleja-Cabre, S., Karimian, A., Joseph, P., and Parry, A., “Investigation into the mechanisms
of propeller-wing interaction noise,” 28th AIAA/CEAS Aeroacoustics 2022 Conference, 2022, p. 2936.

[18] Turhan, B., Kamliya Jawahar, H., Bowen, L., Rezgui, D., and Azarpeyvand, M., “Turbulent Flow Impact on the Acoustic and
Aerodynamic Performance of Overlapping Propellers,” 30th AIAA/CEAS Aeroacoustics Conference, 2024.

[19] Huang, G., Sharma, S., Ambrose, S., and Jefferson-Loveday, R., “Numerical Study on the Aeroacoustics of a Propeller-Wing
Integration at Various Advance Ratios,” Turbo Expo: Power for Land, Sea, and Air, Vol. Volume 1: Aircraft Engine, American
Society of Mechanical Engineers, Boston, Massachusetts, US, 2023, p. V001T01A024.

[20] Turhan, B., Jawahar, H. K., Bowen, L., Rezgui, D., and Azarpeyvand, M., “AEROACOUSTIC CHARACTERISTICS OF
SINGLE PROPELLER-WING CONFIGURATION,” The 29th International Congress on Sound and Vibration, International
Institute of Acoustics and Vibration, Prague, Czech Republic, 2023.

17
 [21] Turhan, B., Kamliya Jawahar, H., Gautam, A., Syed, S., Vakil, G., Rezgui, D., and Azarpeyvand, M., “Acoustic Characteristics
of Phase-Synchronized Adjacent Propellers,” The Journal of the Acoustical Society of America, Vol. accepted, 2024.

[22] Turhan, B., Kamliya Jawahar, H., Gautam, A., Rezgui, D., and Azarpeyvand, M., “Acoustic Performance of Co- and
Counter-Rotating Synchronized Propellers,” 30th AIAA/CEAS Aeroacoustics Conference, 2024.

[23] Shukla, D., and Komerath, N., “Multirotor drone aerodynamic interaction investigation,” Drones, Vol. 2, No. 4, 2018, p. 43.

[24] Schiller, N. H., Pascioni, K. A., and Zawodny, N. S., “Tonal noise control using rotor phase synchronization,” Vertical Flight
Society Annual Forum and Technology Display (VFS Forum 75), 2019.

[25] Guan, S., Lu, Y., Su, T., and Xu, X., “Noise attenuation of quadrotor using phase synchronization method,” Aerospace Science
and Technology, Vol. 118, 2021, p. 107018.

[26] Pascioni, K., and Rizzi, S. A., “Tonal noise prediction of a distributed propulsion unmanned aerial vehicle,” 2018 AIAA/CEAS
Aeroacoustics Conference, Atlanta, Georgia, US, 2018, p. 2951.

[27] Mayer, Y. D., Jawahar, H. K., Szőke, M., Ali, S. A. S., and Azarpeyvand, M., “Design and performance of an aeroacoustic wind
tunnel facility at the University of Bristol,” Applied Acoustics, Vol. 155, 2019, pp. 358–370.

[28] Shao, M., Lu, Y., Xu, X., Guan, S., and Lu, J., “Experimental study on noise reduction of multi-rotor by phase synchronization,”
Journal of Sound and Vibration, Vol. 539, 2022, p. 117199.

[29] Welch, P., “The use of fast Fourier transform for the estimation of power spectra: a method based on time averaging over short,
modified periodograms,” IEEE Transactions on audio and electroacoustics, Vol. 15, No. 2, 1967, pp. 70–73.

[30] Winckelmans, G. S., Topics in vortex methods for the computation of three-and two-dimensional incompressible unsteady flows,
California Institute of Technology, 1989.

[31] Winckelmans, G., and Leonard, A., “Contributions to vortex particle methods for the computation of three-dimensional
incompressible unsteady flows,” Journal of Computational Physics, Vol. 109, No. 2, 1993, pp. 247–273.

[32] Alvarez, E. J., and Ning, A., “Development of a vortex particle code for the modeling of wake interaction in distributed
propulsion,” 2018 Applied Aerodynamics Conference, 2018, p. 3646.

[33] Farassat, F., “Derivation of Formulations 1 and 1A of Farassat,” Tech. rep., 2007.

[34] Pascioni, K. A., Rizzi, S. A., and Schiller, N., “Noise reduction potential of phase control for distributed propulsion vehicles,”
AIAA Scitech 2019 Forum, San Diego, California, US, 2019, p. 1069.

[35] Zarri, A., Koutsoukos, A., Avallone, F., de Prenter, F., Ragni, D., and Casalino, D., “Aeroacoustic Interaction Effects of Adjacent
Propellers in Forward Flight,” 2023.

[36] Casalino, D., Grande, E., Romani, G., Ragni, D., and Avallone, F., “Definition of a benchmark for low Reynolds number
propeller aeroacoustics,” Aerospace Science and Technology, Vol. 113, 2021, p. 106707.

[37] Kunz, F., Zhou, B. Y., Pullin, S. F., Azarpeyvand, M., Galimberti, L., Morelli, M., and Guardone, A., “High-Fidelity Propeller
Broadband Noise Prediction using SU2,” AIAA AVIATION 2023 Forum, 2023, p. 4185.

18

View publication stats