biomimetics

Article
A UDF-Based Approach for the Dynamic Stall Evaluation of
Airfoils for Micro-Air Vehicles
Diana-Andreea Sterpu 1 , Daniel Măriut, a 2 and Lucian-Teodor Grigorie 1, *

1 Faculty of Aerospace Engineering, POLITEHNICA National University of Science and Technology Bucharest,
060042 Bucharest, Romania; [email protected]
2 Department of Aircraft Integrated Systems and Aviation, Military Technical Academy “Ferdinand I”
Bucharest, 050141 Bucharest, Romania; [email protected]
* Correspondence: [email protected]

Abstract: A numerical method for generating dynamic stall using ANSYS Fluent and a user-defined
function (UDF), with the complete script shared for reference, is introduced and tested. The study
draws inspiration from bird flight, exploring dynamic stall as a method for achieving enhanced aero-
dynamic performance. The numerical method was tested on NACA 0012 airfoils with corresponding
chord lengths of c1 = 40 mm, c2 = 150 mm, and c3 = 300 mm at Reynolds numbers ranging from
Re1 = 2.8 × 104 up to Re5 = 1.04 × 106 . Airfoil oscillations were settled for all cases at ω = 0.55 Hz.
Detached eddy simulation (DES) is employed as the turbulence model for the simulations presented,
ensuring the accurate representation of the flow characteristics and dynamic stall phenomena. The
study provides a detailed methodology, encouraging further exploration by researchers, especially
young academics and students.

Keywords: aerodynamics; hysteresis; NACA 0012; DES

1. Introduction
The flight of birds has inspired research and experiments in aerodynamics since the
Citation: Sterpu, D.-A.; Măriut, a, D.; first human attempt at air transportation, successfully achieved by the pioneers of the
Grigorie, L.-T. A UDF-Based first practical airplane, the Wright brothers [1–4]. Approaches to aerodynamic imitation of
Approach for the Dynamic Stall bird flight vary from thrust generation by airfoil oscillations [3] to state-of-the-art shape
Evaluation of Airfoils for Micro-Air morphing methods [4–8].
Vehicles. Biomimetics 2024, 9, 339. One method that is explored to fulfill the human aspiration of soaring like birds
https://doi.org/10.3390/ involves the examination of dynamic stall and its associated advantages. Dynamic stall,
biomimetics9060339 a phenomenon captivating aerodynamic research for over five decades, involves rapidly
Academic Editor: Stanislav N. Gorb changing a wing’s angle of attack (AoA) to create a high lift moment. Thus, dynamic
stall is the aerodynamic phenomenon used by long-distance flying birds to achieve
Received: 12 April 2024 great aerodynamic performance in a short period of time by rapidly changing the AoA
Revised: 20 May 2024 of their wings in order to create high lift moments. This research aims to bridge past
Accepted: 30 May 2024
efforts with contemporary accessibility, emphasizing the importance of a practical and
Published: 4 June 2024
reproducible approach.
Dynamic stall was investigated in 1977 through experiments conducted by National
Aeronautics and Space Administration (NASA) [6,9], which also introduced a software
Copyright: © 2024 by the authors.
package to advance the research further. This research provided empirical data and insights
Licensee MDPI, Basel, Switzerland. into the phenomenon and encouraged further study [6,10,11]. While acknowledging the
This article is an open access article importance of past research, software packages, and methods, it is imperative to embrace
distributed under the terms and a contemporary and accessible approach for replicating dynamic stall. This is crucial
conditions of the Creative Commons to inspire and engage young academics and students, encouraging exploration of the
Attribution (CC BY) license (https:// advantages of dynamic stall and potentially uncovering new insights. The progress of
creativecommons.org/licenses/by/ dynamic stall research has been closely intertwined with the progress in the computational
4.0/). fluid dynamics (CFDs) commercial tools thatwha are available today (e.g., ANSYS Fluent,

Biomimetics 2024, 9, 339. https://doi.org/10.3390/biomimetics9060339 https://www.mdpi.com/journal/biomimetics

Biomimetics 2024, 9, 339 2 of 13

OpenFOAM, Autodesk CFD, etc.). Recent advancements in numerical solutions for un-
steady inviscid and viscous flows have enabled the detailed visualization of essential flow
processes in numerical wind tunnels. This article provides a detailed reproduction method
for dynamic stall, aiming to enhance accessibility for researchers to a validated approach
and to foster innovation and information sharing within the global research community.
The exploration of dynamic stall not only sheds light on fundamental aerodynamic
phenomena but also holds promise for practical applications, particularly in the realm of
biomimetics. As proposed in [12–14], the integration of pitching and plunging motions
of an airfoil for thrust generation represents a compelling avenue for innovation. This
concept finds application across various domains, including solar energy generation, where
dynamic stall phenomena play a crucial role in optimizing the efficiency of turbine blades.
Additionally, in aerospace engineering, the exploration of dynamic stall in pitching and
plunging motions opens avenues for enhancing the maneuverability and performance
of UAVs and micro-air vehicles (MAVs). Moreover, in wind energy applications, such
as vertical axis wind turbines, dynamic stall behaviors are harnessed to improve energy
capture and overall system efficiency. These diverse applications underscore the versatil-
ity and potential of dynamic stall studies in advancing technological innovation across
different fields.
Various research papers address the dynamic stall investigation using NASA’s OVER-
FLOW 2.3 CFD code for delayed detached eddy simulations (DDESs) [15] or ANSYS Fluent
CFD code for simulating 2D dynamic stall on fluctuating freestream [16] or for analyzing
multi-frequency excitation [17]. Additionally, a variety of experimental approaches have
been employed to achieve the same objective [18–20]. Predicting airfoil stalling dynamics
using numerical solutions [21], combining pitch-plunge oscillations of an airfoil for power
production [14], dynamic stall control by airfoil deformation [22], and studying the noise
produced on an oscillating airfoil [23] all contribute to the potential applications of dynamic
stall in both research and industrial contexts. The main focus regarding recent dynamic
stall investigations is placed on rotating wings with helicopter applications [24,25] and
vertical axis wind turbine applications [26,27].
The main objective of this work is testing a user-defined function in Ansys Fluent to
replicate dynamic stall on airfoils suitable for MAVs at relatively high Reynolds numbers.
The novelty lies in a more technical approach for elucidating the process of reproducing
dynamic stall in ANSYS Fluent with the intention of fostering further studies in this field.
In adopting a more practical approach, the intention is to share a tested numerical method,
allowing for immediate utilization and potential improvement by researchers for various
applications, thereby enhancing the accessibility and applicability of dynamic stall studies
in the field of aerodynamics.

2. Methodology
The research design involves dynamic stall testing on a range of airfoil geometries
utilizing the provided UDF script. By selecting the NACA 0012 and sample-sized airfoils,
an easy comparison with existing experimental and numerical results within the research
community is facilitated. The objective is to showcase the alignment of the UDF method
presented in this paper with findings from prior experimental and numerical investigations.
The dissemination of this novel method aims to enhance accessibility to dynamic stall
benefits among academics and foster exploration of various suggested applications outlined
later in the paper.

2.1. Geometry and Mesh Generation

For the current study, NACA 0012 airfoils were selected, with chord lengths of
c1 = 40 mm, c2 = 150 mm, and c3 = 300 mm respectively. The initial geometry places
each airfoil at 0◦ AoA to initiate dynamic stall. The center of gravity (CG), positioned at
chord on the mean camber line of each airfoil, represents an essential aspect of the UDF,
particularly regarding the initiation of motion. Unstructured meshes, employing triangular
Biomimetics 2024, 9, x FOR PEER REVIEW 3 of 13

Biomimetics 2024, 9, 339 airfoil at 0° AoA to initiate dynamic stall. The center of gravity (CG), positioned3at ½
of 13
chord on the mean camber line of each airfoil, represents an essential aspect of the UDF,
particularly regarding the initiation of motion. Unstructured meshes, employing triangu-
lar elements,
elements, werewere generated
generated to to encompass
encompass the
the respectiveairfoil
respective airfoilgeometries
geometries(see
(seeFigure
Figure 1).
Internal zones
Internal zones were
were specifically
specifically established
established near
near each
each airfoil
airfoil to
to capture
capture aa more
moreintricate
intricate
perspectiveof
perspective ofthe
thepressure
pressuredistribution.
distribution.

(a) (b)
Figure1.1. Mesh
Figure Meshof
ofNACA
NACA0012
0012airfoil:
airfoil: (a)
(a) internal
internalzone;
zone;(b)
(b)inflation
inflationlayers.
layers.

Meshing
Meshing is is itself
itself aa preliminary
preliminary stepstep ininaerodynamics
aerodynamics as as well
well as
as an
aninterpretation
interpretation of of
the
the studied
studied effect.
effect. Critical
Critical aspects
aspects ofof this
this stage
stage involve
involve defining
defining the
the study’s
study’s dimensions,
dimensions,
considering
considering the the impact
impact of of walls
walls on
on the
the flow,
flow,ensuring
ensuringthat thatthe
themesh
meshquality
qualityisisreflected
reflectedin in
the obtained results, and determining the dimensions of the internal zone.
the obtained results, and determining the dimensions of the internal zone. Inflation layers Inflation layers
serve
serveasasaavaluable
valuabletool toolfor
forcontrolling
controlling thethe
wall spacing
wall spacingand ensuring
and ensuring appropriate
appropriate y+ values,
y+ val-
where
ues, where y+ represents the nondimensional distance from the wall to the first grid point.
y+ represents the nondimensional distance from the wall to the first grid point.
In
In the
thesimulations
simulations conducted
conducted for for all
allthe
theNACA
NACA0012 0012airfoil
airfoilgeometries,
geometries,including
including chord
chord
lengths
lengths of of c𝑐1 == 40
40mm,mm, c𝑐2 = = 150
150mm,
mm, and and c𝑐3 == 300
300mm,mm, employing
employing fivefive inflation
inflation layers
layers
generated
generatedfrom fromthetheairfoil
airfoilyielded
yieldedsatisfactory
satisfactoryresults,
results,effectively
effectivelycontrolling
controllingy+ y+values
valuesand
and
ensuring accurate boundary layer
ensuring accurate boundary layer resolution. resolution.
Considering
Consideringthe theturbulence
turbulencemodelmodeldesired
desiredfor forreplicating
replicatingdynamic
dynamicstallstallininthis
thispaper
paper
(DES),
(DES), it is necessary to impose careful mesh refinement to adequately resolve both
it is necessary to impose careful mesh refinement to adequately resolve both the
the
near-wall
near-wall regions
regions and and thethe larger
largerturbulent
turbulentstructures.
structures.Inadequate
Inadequate mesh
mesh resolution
resolution cancan
re-
result in numerical inaccuracies, particularly in regions where the flow
sult in numerical inaccuracies, particularly in regions where the flow transitions between transitions between
the
the laminar
laminar and and turbulent
turbulentstates.
states.This
This sensitivity
sensitivity to mesh
to mesh resolution
resolution can make
can make DESsDESscom-
computationally expensive, especially for simulations involving complex geometries or
putationally expensive, especially for simulations involving complex geometries or high
high Reynolds number flows.
Reynolds number flows.
2.2. User-Defined Function
2.2. User-Defined Function
A UDF is a C written script compatible with ANSYS Fluent. In the presented case
study,Athe UDF
UDFisshould
a C written
swiftlyscript
rotatecompatible
the airfoil towith ANSYS
induce Fluent.
dynamic In the
stalling presented
benefits, case
followed
study,
by the UDF should
a recalibration swiftlyAoA.
to the initial rotate the airfoil to induce dynamic stalling benefits, fol-
lowed by a recalibration to the initial
In this specific context, the airfoil AoA. is regarded as a rigid body, while the mesh is
In this specific context, the
the deformable structure. Hence, in eachairfoil is regarded
iteration,asthe
a rigid
meshbody,
needswhile the mesh
to adapt to theis up-
the
deformable structure. Hence, in each iteration, the mesh needs to adapt to
dated position of the airfoil. The tracking of iterations is time-dependent, rendering this the updated
position
study of the airfoil. The tracking of iterations is time-dependent, rendering this study
unsteady.
unsteady.
2.2.1. Mathematical Approach
2.2.1.The
Mathematical
UDF in thisApproach
study serves as the digital mechanism for inducing dynamic stall
TheThis
benefits. UDFcomputational
in this study serves as the
tool lays thegroundwork
digital mechanism for inducing
for delving dynamic stall
into the mathematical
benefits. This computational
underpinnings governing airfoil tool lays the groundwork
dynamics. Within this formathematical
delving into the mathematical
framework, the
variation of the airfoil’s
underpinnings governingAoA between
airfoil a minimum
dynamics. Withinvalue and a maximum
this mathematical value is defined,
framework, the var-
establishing
iation of thea airfoil’s
foundationAoAforbetween
analysis.a This systematic
minimum valueapproach enables the
and a maximum elucidation
value of
is defined,
the intricate dynamics of dynamic stall phenomena with engineering rigor and
establishing a foundation for analysis. This systematic approach enables the elucidation clarity. The
variation of the dynamics
of the intricate airfoil AoA ofisdynamic
definedstall
between a minimum
phenomena with value (αmin ) and
engineering rigora and
maximum
clarity.
value (αmax ):
∆α = αmax − αmin . (1)
The governing pitching of the airfoil equation must be defined as a time-dependent
sinusoidal equation that guarantees the AoA oscillation. For enhanced comprehension, the
 The variation of the airfoil AoA is defined between a minimum value (𝛼 ) and a maxi-
mum value (𝛼 ):
∆𝛼 = 𝛼 𝛼 . (1)
Biomimetics 2024, 9, 339 4 of 13
The governing pitching of the airfoil equation must be defined as a time-dependent
sinusoidal equation that guarantees the AoA oscillation. For enhanced comprehension,
the following scheme represents the transition between the angular velocity 𝜔, defining
following scheme represents the transition between the angular velocity ω, defining circular
circular motion, and the angular frequency 𝜔 = 0.55 Hz, characterizing the simple har-
motion, and the angular frequency ω = 0.55 Hz, characterizing the simple harmonic motion
monic motion (SHM) (see Figure 2).
(SHM) (see Figure 2).

(a) (b)
Figure
Figure 2. Oscillating
2. Oscillating motion:
motion: (a) circular
(a) circular motion;
motion; (b) simple
(b) simple harmonic
harmonic motion
motion.

Following
Following the above
the above scheme,
scheme, point
point A1, A1, defines
which which an
defines an element
element in themo-
in the circular circular
tion, represents the amplitude determination of point A2, which belongs to the SHM. TheSHM.
motion, represents the amplitude determination of point A2, which belongs to the
The following
following statementsstatements can in
can be made be comparing
made in comparing
the abovethe above representations:
representations:
• • circular
In In circular
motion, 𝜔 represents
motion, ω represents the change
the change of angular
of angular displacement;
displacement; in SHM, in SHM, ω
ω rep-
represents
resents thechange
the rate of rate of of
change of theangle.
the phase phase angle.
• • circulation
The The circulation motion’s
motion’s radius 𝑅 is the
radius R isSHM
the SHM amplitude:
amplitude:
𝑦 = 𝑅 sin 𝜔𝑡. (2)
y = Rsin ωt. (2)
• The angular velocity, ω, of the circular motion is also the angular frequency, ω, of the
• The angular velocity, ω, of the circular motion is also the angular frequency, ω, of
SHM.
• the SHM.
The angular frequency (ω) represents how many cycles per second are completed:
• The angular frequency (ω) represents how many cycles per second are completed:
2𝜋
𝜔 = , 2π (3)
ω𝑇 = , (3)
where 𝑇 represents the period (oscillation time). T
where
For thisT paper,
represents the period frequency
the oscillation (oscillationistime).
𝜔 = 0.55 Hz, a value which is grounded
For
in practical this paper,and
applications the aligns
oscillation
withfrequency
biomimetic ω = 0.55 Hz,
is principles a value
observed in which
naturalisflight.
grounded
In biomimetics, where engineering solutions are inspired by biological systems, such fre-flight.
in practical applications and aligns with biomimetic principles observed in natural
In biomimetics,
quencies whereresonant
mimic the natural engineering solutions
frequencies are inspired
observed byand
in birds biological
insects, systems,
ensuring such
frequencies mimic the natural resonant frequencies observed in birds
a closer emulation of their flight dynamics. Through the integration of biomimetics and and insects, ensuring
aerodynamics, the frequency selection in the present study exemplifies the synergistic re- and
a closer emulation of their flight dynamics. Through the integration of biomimetics
aerodynamics,
lationship the frequency
between biological selection
inspiration andin the present
engineering study exemplifies
innovation, paving thetheway
synergistic
for
relationship between biological inspiration
the development of more efficient and agile aerial systems. and engineering innovation, paving the way
for the development of more efficient and agile aerial systems.
2.2.2. Programmable Implementation
2.2.2. Programmable Implementation
In its current state, the provided script (see Table 1) produces half of a dynamic stall,
In its current state, the provided script (see Table 1) produces half of a dynamic stall,
covering the motion from 0° to 18° AoA. This selection is based on extensive aerodynamic
covering the motion from 0◦ to 18◦ AoA. This selection is based on extensive aerodynamic
analyses, considering the critical transitional range where dynamic stall phenomena mani-
fest prominently. The chosen angles of attack are strategically aligned with aerodynamic
principles governing lift and flow separation, ensuring the script’s relevance to dynamic
stall studies across various airfoil geometries. The linear (vel) and angular (omega) veloci-
ties are returned to ANSYS Fluent by overwriting the arrays. The desired result consists in
an oscillating airfoil which is pitching from 0◦ to 18◦ AoA and back [28,29]. The provided
Biomimetics 2024, 9, 339 5 of 13

UDF code offers several advantages over conventional CFD tools due to its programmabil-
ity and adaptability, particularly in handling the complexity of iterations. Unlike standard
CFD software, which may have limited capabilities for customizing dynamic stall simu-
lations, this UDF allows for fine control of the iterative process. By directly interfacing
with ANSYS Fluent, the code empowers researchers to manipulate thread variables, define
custom conditions, and implement specialized algorithms to accurately model dynamic
stall phenomena. This level of control enables the tailoring of simulations to specific re-
quirements, optimizing computational resources and reducing unnecessary calculations.
Moreover, the code’s logical structure and clarity enhance readability and maintainability,
facilitating efficient troubleshooting and code refinement.

Table 1. Main UDF script for reproducing dynamic stall.

Main UDF Body Additional Comments

CG_MOTION—predefined ANSYS function by six arguments:
- airfoil: the name of the UDF;
DEFINE_CG_MOTION(airfoil, dt, vel, omega, time, dtime) - dt (dynamic thread), vel (linear velocity), omega (angular
velocity), time (current time), dtime (time step) are variables
passed by the ANSYS Fluent solver to the UDF Function.
{
Declaring multiple variables in one statement:
- in_omega: initial value for ω;
- t, time;
float in_omega, t, in_T, Stop, alpha_min, alpha_max, - in_T, initial period;
var_omega; - Stop, maximum time allowed for one complete iteration;
- alpha_min, minimum AoA for the dynamic stall;
- alpha_max, the maximum AoA for the dynamic stall;
- var_omega, variable form of ω.
var_omega = 0.0; Initialization of the variable ω.
in_omega = 0.55; Initialization of the initial parameter ω.
Time allocated by the ANSYS Fluent solver to the UDF under the
t=time;
variable t.
in_T = 2 * pi/in_omega; The variable in_T defines how many seconds a cycle lasts.
The parameter Stop defines the maximum time allowed for one
Stop = (in_T + Start);
complete iteration.
alpha_min = (0) * (pi/180); The dynamic stall starts at 0◦ AoA.
alpha_max = (18) * (pi/180); The dynamic stall stops at 18◦ AoA.
if (t <= Start)
{
var_omega = 0; Exit loop 1, without variable modification.
omega [0] = var_omega;
}
else{
if (t > Start && t <= Stop)
{
var_omega= ((0.5 * in_omega * (alpha_max- alpha_min)) * Exit loop 2, with variable modification.
sin(in_omega * (t-Start)));
omega [0] = var_omega;
}
else{
if (t > Stop) Exit loop 3, without variable modification.
{
var_omega = 0;
omega [0] = var_omega;
}
}
}
}
Biomimetics 2024, 9, 339 6 of 13

Future research endeavors may focus on extending the capabilities of the UDF ap-
proach to encompass additional functionalities, thereby broadening its utility in biomimetic-
inspired aerodynamic investigations. One potential avenue for enhancement lies in inte-
grating features that account for airfoil deformation, thereby enabling more comprehensive
modeling of dynamic stall phenomena with consideration for structural flexibility. Ad-
ditionally, efforts could be directed towards extending the UDF to support dynamic stall
simulations in three-dimensional geometries, facilitating the exploration of complex flow
dynamics and interactions in scenarios more closely resembling real-world conditions
encountered in both biological and industrial contexts. Furthermore, the UDF could be
tailored to simulate aerodynamic interactions in multi-body systems, facilitating investiga-
tions into collective behavior and cooperative flight strategies inspired by social insects or
flocking birds.
The simulations presented in this paper required moderate computational resources,
typically available in standard desktop or workstation environments, facilitating accessibil-
ity and reproducibility.

3. Results and Discussion

3.1. Numerical Results for NACA 0012—c3 = 300 mm
Firstly, the UDF was tested on a NACA 0012 profile of larger chord length (c3 = 300 mm)
at Re5 = 1.04 × 106 . The results are presented in Figure 3 and serve as an example of the
effect of dynamic stall on the pressure contour of the airfoil, juxtaposed with the theoretical
polar curve depicting the lift coefficient against the AoA. The pressure contour on the lower
contour of the airfoil is slightly bigger than the one on the upper contour, resulting in lift
generation (Figure 3c,d). The dynamic-stall vortex is forming at the leading edge of the
airfoil. The swift rotation of the airfoil to a AoA does not immediately induce a pressure
change on the upper surface corresponding to the new AoA. This delay is attributed to the
time lag between the airfoil’s motion and the subsequent pressure response (Figure 3e,f).
The dynamic-stall vortex advanced on the upper contour of the airfoil from the leading
edge to the trailing edge. The airfoil momentarily reacts to a lower AoA than the geometric
AoA at its current state. The delayed response to the geometrical AoA also results in a
postponement of the boundary layer separation. The vortex is initiated near the leading
edge of the airfoil and progresses toward the trailing edge along the upper contour of the
airfoil. During this stage, a substantial additional lift is generated. Once the dynamic stall
vortex surpasses the trailing edge, there is a notable reduction in lift (Figure 3g,h). Upon
reaching the ultimate stage, leading to complete flow separation (deep stall state), the flow
reattachment phase persists as the AoA decreases. The process concludes when the flow is
fully reattached to the contour of the airfoil (Figure 3i,j).
The specific behavior of dynamic stall is highly contingent upon the geometry of the
airfoil and the Reynolds number to which the airfoil is subjected. With that in mind, the
presented results should function as both a guide and a validation of the methodology
outlined in this study.
 the leading edge of the airfoil and progresses toward the trailing edge along the upper
contour of the airfoil. During this stage, a substantial additional lift is generated. Once the
dynamic stall vortex surpasses the trailing edge, there is a notable reduction in lift (Figure
Biomimetics 2024, 9, 339 3g,h). Upon reaching the ultimate stage, leading to complete flow separation (deep 7 of 13
stall
state), the flow reattachment phase persists as the AoA decreases. The process concludes
when the flow is fully reattached to the contour of the airfoil (Figure 3i,j).

(a) (b)

(c) (d)

(e) (f)
Figure 3. Cont.
Biomimetics 2024,
Biomimetics 9,9,x 339
2024, FOR PEER REVIEW 8 of 13
8 of 13

(g) (h)

(i) (j)
Figure
Figure 3.
3. (a)
(a) Pressure contour at
Pressure contour at00° AoA.(b)
◦ AoA. (b)Evolution
Evolutionofof the
the polar
polar curve
curve at 0at 0° AoA.
◦ AoA. (c) Pressure
(c) Pressure
contour
contour at 3.2◦ AoA. (d) Evolution of the polar curve at 3.2◦ AoA. (e) Pressure contour at 9.819.81°
at 3.2° AoA. (d) Evolution of the polar curve at 3.2° AoA. (e) Pressure contour at AoA.
◦ AoA.
(f) Evolution of the polar curve when vortex forms near the leading edge at 9.81° AoA. (g) Pressure
(f) Evolution of the polar curve when vortex forms near the leading edge at 9.81◦ AoA. (g) Pressure
contour at 13.08°◦AoA. (h) Evolution of the polar curve when moment stall occurs at 13.08° AoA. (i)
contour at 13.08 AoA. (h) Evolution of the polar curve when moment stall occurs at 13.08◦ AoA.
Pressure contour at 17.98° AoA (j) Evolution of the polar curve with significant drag increase caused
(i) Pressure contour at 17.98◦ AoA (j) Evolution of the polar curve with significant drag increase
by the separation of the flow.
caused by the separation of the flow.

The specific
3.2. Numerical behavior
Results of dynamic
for NACA 0012—cstall is highly contingent upon the geometry of the
2 = 150 mm
airfoil and the Reynolds number to which the airfoil is subjected. With that in mind, the
In this section, the results obtained from the simulations using the NACA 0012 airfoil
presented results should function as both a guide and a validation of the methodology
with a c2 = 150 mm chord length at Re4 = 5.2 × 105 and Re5 = 1.04 × 106 are presented (see
outlined in this study.
Figures 3 and 4). The investigation reveals that the results obtained through the utilization
of the UDF exhibit behavior consistent with the experimental findings (as compared in
3.2. Numerical
Table Results forpresented
2 and graphically NACA 0012—𝑐 = 4).
in Figure 150Notably,
𝑚𝑚 the uplift curve demonstrates a
In this
higher section, the
lift coefficient results obtained
compared to any offrom the simulations
the other using
experimental the NACA
results. 0012
However, airfoil
such
a 𝑐 =
disparities
with are150 mm chord
commonly encountered
length atand𝑅𝑒deemed
= 5.2 normal
× 10 and 𝑅𝑒 = 1.04analyses.
in aerodynamic × 10 are It ispre-
essential
sented (seeto Figures
note that3the experimental
and results were
4). The investigation conducted
reveals on diverse
that the results geometries and
obtained through
AoAs, which are comprehensively presented in Table 2.
the utilization of the UDF exhibit behavior consistent with the experimental findings (as
compared in Table 2and graphically presented in Figure 4). Notably, the uplift curve
demonstrates a higher lift coefficient compared to any of the other experimental results.
However, such disparities are commonly encountered and deemed normal in aerody-
namic analyses. It is essential to note that the experimental results were conducted on
diverse geometries and AoAs, which are comprehensively presented in Table 2.

Table 2. Comparison of the obtained numerical results against experimental results.

Reynolds
AoA Range
Method 0000 Observations
[mino–maxo]
Number
 1.04 × 10 0°–18° NACA 0012, 𝑐 = 150 mm
Experimental—Reddy et al. [7] 2.5 × 10 5°–25° NACA 0012, 𝑐 = 127 mm
Experimental—Surekha et al. [11] 10 −5°–25° NACA 0012, 𝑐 = 160 mm
Experimental—McAlister et al. [5] 2.5 × 10 0°–20° NACA 0012, 𝑐 = 76.2 mm
Biomimetics 2024, 9, 339 9 of 13

Figure4.4.Comparison
Figure Comparison of the
of the results
results for NACA
for NACA 0012, c0012, 𝑐 = 150 mm of this paper against experi-
2 = 150 mm of this paper against experimental
mentalfrom
results results from
Reddy et Reddy et al.
al. (1987) (1987)
[11]; [11];etSurekha
Surekha al. (2019)et[18];
al. (2019) [18]; andetMcAlister
and McAlister al. (1978) et
[6].al. (1978) [6].

The
Table 2. followingofTable
Comparison 3 and numerical
the obtained Figure 5 results
illustrate theexperimental
against testing of the UDF-obtained results
results.
against the numerical results from other studies. Comparing numerical results is crucial
Reynolds AoA Range
to validate the accuracy of the presented UDF, which
Method is functioning within ANSYS Fluent,
Observations
Number [min◦ –max◦ ]
ensuring that the findings are reliable and
5
consistent
◦
with
◦
established data.
5.2 × 10 0 –18 NACA 0012, c = 150 mm
Present work
1.04 × 106 0◦ –18◦ NACA 0012, c = 150 mm
Table 3. Comparison of obtained results against other CFD results.
Experimental—Reddy et al. [7] 2.5 × 106 5◦ –25◦ NACA 0012, c = 127 mm
Experimental—Surekha et al. [11] 5
10Turbulence −5◦ –25 ◦
AoA Range NACA 0012, c = 160 mm
Method
Experimental—McAlister et al. Re
[5] 2.5 × 106 0◦ –20◦ Observations
NACA 0012, c = 76.2 mm
Model [mino–maxo]
NACA 0012, 𝑐 =
The following Table 35.2
and×Figure
10 DES the testing
5 illustrate 0°–18°
of the UDF-obtained
150 mmresults
Present
against work results from other studies. Comparing numerical results is crucial to
the numerical
NACA 0012, 𝑐 =
1.04
validate the accuracy of the × 10 UDF,DES
presented 0°–18° within ANSYS Fluent,
which is functioning
150 mm
ensuring that the findings are reliable and consistent with established data.
3D NACA 0012, 𝑐 =
CFD—Fan et al. [24] 3.2 × 10 SST gamma 0°–18°
Table 3. Comparison of obtained results against other CFD results.
35 mm
CFD—Baldan et al. NACA 0012, 𝑐 =
1.35 × 10 URANS −5°–25°
[25] AoA Range 150 mm
Method Re Turbulence Model Observations
[min◦ –max◦ ]
5.2 × 105 DES 0◦ –18◦ NACA 0012, c = 150 mm
Present work
1.04 × 106 DES 0◦ –18◦ NACA 0012, c = 150 mm
CFD—Fan et al. [24] 3.2 × 104 SST gamma 0◦ –18◦ 3D NACA 0012, c = 35 mm
CFD—Baldan et al. [25] 1.35 × 105 URANS −5◦ –25◦ NACA 0012, c = 150 mm
 Biomimetics 2024,
Biomimetics 2024, 9,
9, 339
x FOR PEER REVIEW 10 of10
13 of 13

Figure5.5.Comparison
Figure Comparisonbetween
between current
current work
work results
results forfor NACA
NACA c2 =c 150
0012,
0012, = mm
150 mm and CFD
and CFD results
results
obtained by Fan et al. (2019) [30], and Baldan et al. (2023) [31].
obtained by Fan et al. (2019) [30], and Baldan et al. (2023) [31].

3.3.Numerical
3.3. NumericalResults
Resultsforfor NACA
NACA 0012—𝑐
0012—c = 40 𝑚𝑚
1 = 40 mm
Theexamination
The examination of of
thethec1 =𝑐 40 = mm
40 mm chordchord
lengthlength
NACA NACA 0012 airfoil,
0012 airfoil, as wellas aswell
micro as mi-
cro airfoils
airfoils in general,
in general, is a crucial
is a crucial aspect aspect of aerodynamic
of aerodynamic research,
research, particularly
particularly concerning
concerning
MAVs,
MAVs,drones,
drones,and andother
other small-scale
small-scale aerial platforms.
aerial TheseThese
platforms. airfoilsairfoils
play a play
vital role in de-
a vital role in
termining key factors like maneuverability, stability, and efficiency.
determining key factors like maneuverability, stability, and efficiency. Figure 6 illustrates Figure 6 illustrates the
outcomes
the outcomesobtained from from
obtained the examination
the examination of the of
NACA 0012 airfoil
the NACA 0012 with a chord
airfoil with alength
chord of length
cof= 40 mm undergoing dynamic stall under various
1 𝑐 = 40 mm undergoing dynamic stall under various Reynolds 1numbers, Reynolds numbers, Re = 2.8 × 104𝑅𝑒 , =
4 5
Re 2 =
2.8 10 ×, 10
× 8.4 = 8.4Re×
𝑅𝑒 , and 3 = 101.4 × 10 𝑅𝑒
, and . Additionally,
= 1.4 × 10 drawing inspiration
. Additionally, from the
drawing effi-
inspiration
cient aerodynamics observed in small flying creatures such as insects or birds contributes
from the efficient aerodynamics observed in small flying creatures such as insects or birds
valuable insights to the design and improvement of micro airfoils used in MAVs or drones.
contributes valuable insights to the design and improvement of micro airfoils used in
Analyzing Figure 6, it can be observed that the lower the Reynolds number, the lower
MAVs or drones.
the lift slope is. At Re1 , it can be observed that the lift coefficients are close to zero for the
returnAnalyzing Figurelift
line. The highest 6, itslope
can be observedatthat
is obtained Re3the lower ifthe
. Overall, theReynolds number,
lift coefficients from thealllower
the lift slope is. At 𝑅𝑒 , it can be observed that the lift coefficients
the three cases are analyzed, it can be observed that lift coefficients decrease when the chord are close to zero for the
returnisline.
length The highest
decreased. lift slope is DES
By incorporating obtained at 𝑅𝑒
into the . Overall,
numerical if the liftof
simulations coefficients from all
micro airfoils,
this study enhances the fidelity and accuracy of the results obtained. Specifically, DES the
the three cases are analyzed, it can be observed that lift coefficients decrease when
chord length
enables is decreased.
the capture By incorporating
of both near-wall turbulenceDES andinto the numerical
large-scale unsteadysimulations
flow structures, of micro
airfoils, this
providing study enhances
a comprehensive the fidelity of
representation andtheaccuracy
complex of the resultsinteractions
aerodynamic obtained. Specifically,
during
dynamic
DES enablesstall. the capture of both near-wall turbulence and large-scale unsteady flow struc-
tures, providingdepicted
The results in Figure 6representation
a comprehensive serve as a testamentof the to the reliability
complex and versatility
aerodynamic interactions
of the method employed,
during dynamic stall. particularly the utilization of the user-defined function (UDF)
acrossThe
varying
results Reynolds
depicted numbers.
in Figure Through
6 servethisas acomprehensive
testament to the analysis, encompassing
reliability and versatility
different
of the method employed, particularly the utilization of the user-defined function of
Reynolds numbers and corresponding aerodynamic phenomena, the efficacy (UDF)
the proposed approach is underscored.
across varying Reynolds numbers. Through this comprehensive analysis, encompassing
different Reynolds numbers and corresponding aerodynamic phenomena, the efficacy of
the proposed approach is underscored.
 Biomimetics 2024,
Biomimetics 2024, 9,
9, 339
x FOR PEER REVIEW 11 of11
13 of 13

Figure6.6.Polar
Figure Polarcurve
curveofofNACA
NACA 0012
0012 for for 𝑐 40
c1 = = 40
mmmm 𝑅𝑒2.8
at Reat1 = =×2.8
10×
4 , 10 𝑅𝑒8.4=×8.4
Re2, = 104×, and
10 , and
𝑅𝑒 = 1.4 × 10
Re3 = 1.4 × 105 . .

4.4.Conclusions
Conclusions
Byproviding
By providinga acomprehensive
comprehensive exposition
exposition of the
of the exact
exact methodology
methodology for reproducing
for reproducing
dynamicstall
dynamic stallononany
anyairfoil
airfoil using
using a UDF
a UDF andand ANSYS
ANSYS Fluent,
Fluent, the the primary
primary objective
objective of this
of this
paper is to facilitate a more accessible approach for venturing into the fields
paper is to facilitate a more accessible approach for venturing into the fields of aerody- of aerodynamics
and dynamic
namics stall research.
and dynamic stall research.
In
Inanalyzing
analyzingthe theresults
results obtained
obtained using thethe
using presented
presentedmethodmethodand and
juxtaposing themthem
juxtaposing
with findings from the literature, it becomes evident that variations
with findings from the literature, it becomes evident that variations in the turbulence in the turbulence
model,
model,airfoil size,
airfoil range
size, rangeof AoA,
of AoA,andandmesh quality
mesh contribute
quality significantly
contribute to discrepancies
significantly to discrepan-
between results. Nonetheless, upon closer examination, it is apparent that despite these
cies between results. Nonetheless, upon closer examination, it is apparent that despite
discrepancies, all CFD models compared cover similar areas of results. This observation
these discrepancies, all CFD models compared cover similar areas of results. This obser-
underscores the robustness and effectiveness of the method outlined in this paper.
vation underscores the robustness and effectiveness of the method outlined in this paper.
The incorporation of the UDF presented within ANSYS Fluent not only allows for
Thedegree
a higher incorporation of the UDF
of customization but presented
also enhanceswithin
the ANSYS
fidelity of Fluent
results.notByonly allows for a
leveraging
higher degree of customization but also enhances the fidelity
UDFs, the simulation methodology gains the capability to finely tune parameters, such of results. By leveraging
as
UDFs, the simulation methodology gains the capability to finely tune
frequency, angle of attack range, and airfoil dynamics, resulting in a more tailored approach parameters, such as
frequency,
to investigatingangle of attack
dynamic range,
stall effectsand onairfoil dynamics,
airfoils. This level resulting in a more enables
of customization tailored ap-
proach to investigating
simulations dynamic
to capture intricate flowstall effectswith
behaviors on airfoils. This level
greater detail, of customization
thereby improving the ena-
bles simulations
accuracy to capture
and reliability of theintricate
results. flow
This behaviors with greater
approach allows for thedetail,
seamless thereby improving
coupling of
sophisticated aerodynamic analyses with customizable airfoil kinematics,
the accuracy and reliability of the results. This approach allows for the seamless coupling enabling the
investigation
of sophisticated of complex flow interactions
aerodynamic analyses with with confidence. airfoil
customizable As a result, while the
kinematics, flow the
enabling
calculations
investigation leverage the established
of complex capabilities with
flow interactions of ANSYS Fluent, the
confidence. Ascontrolled movement
a result, while the flow
of the airfoil through the UDF ensures the fidelity and reliability of the
calculations leverage the established capabilities of ANSYS Fluent, the controlled move- simulation outcomes,
ment of the airfoil through the UDF ensures the fidelity and reliability of the simulation
outcomes, providing valuable insights into aerodynamic performance. In essence, the
Biomimetics 2024, 9, 339 12 of 13

providing valuable insights into aerodynamic performance. In essence, the utilization of

UDFs in ANSYS Fluent serves to bolster the robustness and comprehensiveness of the
simulation methodology.
This paper inherently aligns with the principles of biomimetics, as it leverages in-
sights from nature’s efficient flyers, such as birds and insects, to optimize the design and
performance of micro airfoils. By emulating the aerodynamic characteristics observed in
biological systems, this paper aims to set the basis for developing airfoil designs that mimic
natural wing shapes, surface textures, and flight dynamics. Through the investigation of
dynamic stall phenomena, which are prevalent in avian flight, this research endeavors to
apply biomimetic principles to enhance the maneuverability, stability, and efficiency of
MAVs and drones. The integration of biomimetic insights into airfoil design and analysis
represents a significant contribution to the field of bio-inspired engineering, where solu-
tions inspired by nature are utilized to address complex engineering challenges. Overall,
the exploration of dynamic stall in the context of biomimetics and aerodynamics presents
promising opportunities for interdisciplinary research and technological advancement with
broad societal implications.

Author Contributions: Conceptualization, methodology, software, validation, formal analysis, inves-

tigation, resources, data curation, writing—original draft preparation, writing—review and editing,
D.-A.S.; writing—review and editing, visualization, supervision, D.M.; and visualization, supervision,
L.-T.G. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Data Availability Statement: Not applicable.
Conflicts of Interest: The authors declare no conflicts of interest.

References
1. Anderson, J. Fundamentals of Aerodynamics, 6th ed.; McGraw Hill: New York, NY, USA, 2013; ISBN 978-1259129919.
2. Concilio, A.; Dimino, I.; Ameduri, S.; Pecora, R. Adaptive Structures and Biomimetic Robots—A Perspective. Int. J. Robot. Autom.
Technol. 2020, 7, 50–55. [CrossRef]
3. Cebeci, T.; Platzer, M.; Chen, H.; Chang, K.-C.; Shao, J.P. Analysis of Low-Speed Unsteady Airfoil Flows; Horizons Publishing Inc.:
San Jose, CA, USA, 2005.
4. Ricci, S.; Aliabadi, F.M.H.; Botez, R.; Semperlotti, F. Morphing Wing Technologies Large Commercial Aircraft and Civil Helicopters;
Butterworth-Heinemann, Elsevier: Amsterdam, The Netherlands, 2017.
5. Kerho, M. Adaptive airfoil dynamic stall control. J. Aircr. 2007, 44, 1350–1360. [CrossRef]
6. Mcalister, K.W. Dynamic Stall Experiments on the NACA 0012 Airfoil; United States National Aeronautics and Space Administration:
Washington, DC, USA, 1978.
7. Ameduri, S.; Concilio, A. Morphing wings review: Aims, challenges, and current open issues of a technology. Proc. Inst. Mech.
Eng. Part C J. Mech. Eng. Sci. 2023, 237, 4112–4130. [CrossRef]
8. Giuliani, M.; Dimino, I.; Ameduri, S.; Pecora, R.; Concilio, A. Status and Perspectives of Commercial Aircraft Morphing.
Biomimetics 2022, 7, 11. [CrossRef] [PubMed]
9. Carr, L.W.; Mcalister, K.W.; Mccroskey, W.J. Analysis of the Development of Dynamic Stall Based on Oscillating Airfoil Experiments;
United States National Aeronautics and Space Administration: Washington, DC, USA, 1977.
10. Mccroskey, W.J.; Mcalister, K.W.; Carr, L.W.; Pucci, S.L. An Experimental Study of Dynamic Stall on Advanced Airfoil Sections. Volume
1: Summary of the Experiment; United States National Aeronautics and Space Administration: Washington, DC, USA, 1982.
11. Reddy, T.S.R.; Kaza, K.R.V. A Comparative Study of Some Dynamic Stall Models; NASA Lewis Research Center: Cleveland, OH, USA,
1987.
12. Bowker, J.A.; Townsend, N.C.; Tan, M.; Shenoi, R.A. Experimental study of a wave energy scavenging system onboard autonomous
surface vessels (ASVs). In Proceedings of the OCEANS 2015—Genova, Genova, Italy, 18–21 May 2015; IEEE: Piscataway, NJ, USA,
2015; pp. 1–9.
13. Lennie, M.D. Development of the QFEM Solver. Master’s Thesis, Royal Institute of Technology, Stockholm, Sweden, 2013.
14. Tuncer, I.; Walz, R.; Platzer, M. Computational Study on the Dynamic Stall of a Flaping Airfoil. In Proceedings of the 6th AIAA
Applied Aerodynamics Conference, Albuquerque, NM, USA, 15–18 June 1998; p. 2519.
15. Batther, J.; Lee, S. Numerical investigation of a pitching airfoil undergoing dynamic stall using Delayed Detached Eddy
Simulations. Comput. Fluids 2022, 249, 105691. [CrossRef]
Biomimetics 2024, 9, 339 13 of 13

16. Al-Jaburi, K.; Feszty, D.; Nitzsche, F. A methodology for simulating 2D shock-induced dynamic stall at flight test-based fluctuating
freestream. Chin. J. Aeronaut. 2019, 32, 2223–2238. [CrossRef]
17. Huang, D.; Li, J.; Liu, Y. Airfoil Dynamic Stall and Aeroelastic Analysis Based on Multi-frequency Excitation Using CFD Method.
Procedia Eng. 2015, 99, 686–695. [CrossRef]
18. Surekha, R.S.D.; Khandelwal, A.; Rajasekar, R. Investigation of flow field in deep dynamic stall over an oscillating NACA 0012
airfoil. J. Appl. Fluid Mech. 2019, 12, 857–863. [CrossRef]
19. Raghav, V.; Richards, P.; Komerath, N.; Smith, M.J. An Exploration of the Physics of Dynamic Stall. In Proceedings of the AHS
Aeromechanics Specialists Conference, San Francisco, CA, USA, 20–22 January 2010.
20. Wood, J.N.; Breuer, M.; De Nayer, G. Experimental investigations on the dynamic behavior of a 2-DOF airfoil in the transitional
Re number regime based on digital-image correlation measurements. J. Fluids Struct. 2020, 96, 103052. [CrossRef]
21. Adibi, T.; Razavi, S.E.; Ahmed, S.F.; Hassanpour, H.; Saha, S.C.; Muyeen, S.M. Predicting airfoil stalling dynamics using upwind
numerical solutions to non-viscous equations. Results Eng. 2023, 20, 101472. [CrossRef]
22. Geissler, W.; Raffel, M. Dynamic stall control by airfoil deformation. In Proceedings of the Nineteenth European Rotorcraft Forum,
Cernobbio, Italy, 14–16 September 1993.
23. Raus, D.; Cotté, B.; Monchaux, R.; Jondeau, E.; Souchotte, P.; Roger, M. Experimental study of the dynamic stall noise on an
oscillating airfoil. J. Sound Vib. 2022, 537, 117144. [CrossRef]
24. Gardner, A.D.; Jones, A.R.; Mulleners, K.; Naughton, J.W.; Smith, M.J. Review of rotating wing dynamic stall: Experiments and
flow control. Prog. Aerosp. Sci. 2023, 137, 100887. [CrossRef]
25. Wang, Q.; Zhao, Q.; Wu, Q. Aerodynamic shape optimization for alleviating dynamic stall characteristics of helicopter rotor
airfoil. Chin. J. Aeronaut. 2015, 28, 346–356. [CrossRef]
26. Le Fouest, S.; Mulleners, K. The dynamic stall dilemma for vertical-axis wind turbines. Renew. Energy 2022, 198, 505–520.
[CrossRef]
27. Florin, F.; Horia, D.; Alexandru, D. Numerical Investigations of Dynamic Stall Control. INCAS Bull. 2014, 6, 67–80. [CrossRef]
28. udf.h Header File, ANSYS Inc. Available online: https://www.afs.enea.it/project/neptunius/docs/fluent/html/udf/node8.htm
(accessed on 15 January 2024).
29. Define CG_MOTION, ANSYS Inc. Available online: https://www.afs.enea.it/project/neptunius/docs/fluent/html/udf/node8
3.htm (accessed on 18 January 2024).
30. Fan, C.; Wu, Y.; Yu, Y.; Wang, A. Dynamic Thin-Airfoil Stall and Leading-Edge Stall of Oscillating Wings at Low Reynolds
Number. Fluid Dyn. 2019, 54, 691–704. [CrossRef]
31. Baldan, G. Pattern recognition of the flow around a pitching NACA 0012 airfoil in dynamic stall conditions. In Materials Research
Proceedings; Materials Research Forum LLC: Millersville PA, USA, 2023; pp. 161–165.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.