Progress in Aerospace Sciences 110 (2019) 100541

Contents lists available at ScienceDirect

Progress in Aerospace Sciences

journal homepage: www.elsevier.com/locate/paerosci

Models for wind tunnel tests based on additive manufacturing technology T

a,b,∗
Weijun Zhu
a
Collaborative Innovation Center of High-end Manufacturing Equipment, Xi'an Jiaotong University, 710049, Xi'an, Shaanxi, People's Republic of China
b
State Key Laboratory for Manufacturing Systems Engineering, Xi'an Jiaotong University, 710049, Xi'an, Shaanxi, People's Republic of China

ARTICLE INFO ABSTRACT

Keywords: Wind tunnel testing is a reliable means for aircraft design. The wind tunnel models are the objects used in the
Wind tunnel models tests. The accuracy and economy of the model design and fabrication have an important impact on the quality
Wind tunnel tests and cycle of aircraft development. Additive Manufacturing (AM, or Rapid Prototyping, 3D printing) can directly
Additive manufacturing fabricate 3D parts through accumulating raw materials, and is widely regarded as a revolutionary advancement
3D printing
in manufacturing technology. In the very early development period, AM was soon introduced and was studied by
Aerospace
Aircraft
many groups worldwide. Firstly, the introduction of AM is an advancement for the fabrication of models, which
Air vehicle can greatly improve the fabrication economy of current models, such as reducing the number of parts, and
shortening the processing cycle etc. Secondly, the introduction of AM can also improve the design of models,
which is helpful to develop new types of models and even new test methods. Thirdly, AM has blurred the
boundaries between real aircraft and experimental models, and promoted the development of new concept
aircraft. The review first introduces the design requirements of the wind tunnel test models and recalls the AM
application history for models. Next, detailed technologies concerning the design procedure and fabrication
processing of the AM-based models are presented. Finally, the application of AM-based model in the develop-
ment of current air vehicles and new-concept vehicles is introduced. The review provides an overview of
techniques of AM in wind tunnel test models, and provide typical examples for reference to designers and
researchers in the aerospace industry.

1. Introduction control surfaces, inlets, nozzles, etc.) as opposed to more simple reentry
capsules and aerobrakes [1–4,6,7]. The continued heavy reliance on
1.1. Wind tunnel tests and models wind tunnel testing can be seen in Fig. 1. The increased complexity and
scope of air vehicle designs, coupled with the need to avoid risky and
Wind tunnel testing, together with flight testing and numerical costly problem identification during flight testing, have actually driven
computing, is indispensable for aerodynamics research and air vehicle continued increases in wind tunnel testing [8].
development [1–4]. The flight testing uses the real aircraft to operate in Wind tunnel models are reduced-scale alternatives to the prototypes
the real atmosphere, and the obtained data has the strongest reliability. of aerial vehicles used in wind tunnels. Therefore, the creation of many
As the test conditions are difficult to control and change, however, the wind tunnel models is usually involved in the development of a new air
obtained data is highly dispersive, and it is difficult to establish the vehicles [9,10]. The design and fabrication of wind-tunnel models has
intrinsic law of physical phenomena. The solving capacity and effi- traditionally been a highly skilled and time-consuming process [11],
ciency of numerical computing, represented by CFD, is continuously and it should meet similarity requirements, measuring requirements,
improved. It can replace some wind tunnel testing, and can provide strength requirements and processing requirements [12–14]. In a ty-
aerodynamic data more quickly and economically [5]. However, for the pical test, the surface profile dimensional tolerance and surface finish
complex air flow phenomenon often encountered in the development of are set to be within ± 0.1 mm and 0.8 to 3.2 RMS in micrometers, to
advanced air vehicles, its reliability encounters great challenges. De- eliminate the influence on the stability of the boundary layer during
spite the change in emphasis and purpose, wind tunnel testing remains testing [10]. During wind tunnel tests, mechanical stress is usually
the largest and most extensive source of data for major programs. This present on wind tunnel model components. For some components,
is particularly true for advanced aerospace vehicles having complex when the wind speed increases, the stress increases with speed squared
geometries such as X-33, X-34 and Hyper-X (e.g., vehicles having fins, [15].

∗
Collaborative Innovation Center of High-end Manufacturing Equipment, Xi'an Jiaotong University, 710049, Xi'an, Shaanxi, People's Republic of China.
E-mail address: [email protected].

https://doi.org/10.1016/j.paerosci.2019.05.001
Received 20 January 2019; Received in revised form 3 May 2019; Accepted 9 May 2019
Available online 16 May 2019
0376-0421/ © 2019 Elsevier Ltd. All rights reserved.
W. Zhu Progress in Aerospace Sciences 110 (2019) 100541

Nomenclature RT rapid tooling

3DP three-dimensional printing
α, AoA angle of attack, degree (°) SL stereolithography
CFD computational fluid dynamics FDM fused deposition modeling
CSD computational structural dynamics SLS selective laser sintering
CAD computer aided design SLM selective laser melting
FEA finite element analysis CNC computer numerical control
EFD experimental fluid dynamics LOM laminated object manufacturing
AM additive manufacturing STL stereolithography
RP rapid prototyping AMF additive manufacturing file

Fig. 1. History of wind tunnel testing on major aircraft programs [8].

Traditional wind tunnel models are constructed of metal for high- 1.2. Additive manufacturing (AM)
speed testing, with fiberglass, foam, or wood added to the mix of ma-
terials for low-speed testing. These construction methods are frequently AM is an “additive” process which combines layers of paper, wax,
time consuming and costly requiring long lead times in order to execute plastic, or resin to create a solid object [23,24]. AM is also known as
model fabrication for a test program. To better respond to future air- Rapid Prototyping (RP), 3D Printing, etc. This technology is different
craft design processes, current methods of wind tunnel model fabrica- from most machining processes which are “subtractive,” such as mil-
tion must be improved to enable a test program to be executed more ling, drilling, grinding, etc. that remove material from a solid block to
rapidly. With today's CAD and CFD capabilities, aircraft design concepts form an object. This allows the creation of complicated objects with
are being evaluated and discarded in one-third of the time that it takes internal features that cannot be manufactured by other means at low
to construct a typical model. This puts wind tunnel testing in a lagging cost, making it an ideal process for aerospace applications (and models
position, solely to validate predictions. To reduce this time to weeks, in particular). AM is a class of technologies that automates the physical
even days, will require a revolutionary change in the way models are construction directly from a CAD database. These “three-dimensional
designed and fabricated [6]. The possibilities of speeding up the process printers” quickly create tangible parts for purposes from display to test
of constructing aerodynamic models with conventional technology articles. The aerospace industry has adapted a number of these tech-
based on machining their components made of steel and aluminum nologies such as stereolithography (SL), selective laser sintering (SLS),
alloys are considerably limited [16]. Tools such as AM and high-speed and fused deposition modeling (FDM) to construct inexpensive models
machining are being used to significantly reduce both the cost and time for primarily low-speed testing or, in more limited cases, for high-speed
required for model fabrication [9]. The AM appeared to be the best testing. The processes that are used for model fabrication and their
method for manufacturing a model [17–20]. In addition to a possible ASTM-approved terms are listed in the following [24]:
sharp decrease in time and costs of producing force measurement
models meeting geometrical requirements, the new technology allows a ∙ Material extrusion—an additive manufacturing process in which
significant decrease in the time required for construction of models material is selectively dispensed through a nozzle or orifice, com-
with elastically and dynamically similarity [18,21,22]. monly known as FDM;

2
W. Zhu Progress in Aerospace Sciences 110 (2019) 100541

∙ Material jetting—an additive manufacturing process in which [26,28,29].

droplets of build material are selectively deposited, commonly In this review, only models for aircraft used in wind tunnel testing
known as 3D Printing; are included. Other AM-based models, as shown in Fig. 2, are not dis-
∙ Binder jetting—an additive manufacturing process in which a li- cussed, e.g. buildings [30–33], bridges [34], automobiles [35–37]、
quid bonding agent is selectively deposited to join powder materials, sports apparatus [38]、wind blades [39,40]、sea vehicles [41]. Models
commonly known as Polyjet; of non-aerodynamic structures of aircraft are not included, e.g. engines
∙ Sheet lamination—an additive manufacturing process in which [42–44], air particle separators [45], rotary wings [46]. Models of
sheets of material are bonded to form an object, commonly known aircraft used for other purposes are also not discussed, e.g. functional
as LOM; show [46,47], education [31,48] and water tunnel testing [49].
∙ Vat photopolymerization—an additive manufacturing process in Over the past two decades, multiple groups around the world have
which liquid photopolymer in a vat is selectively cured by light- used a variety of AM processes and AM materials to design and fabri-
activated polymerization, commonly known as SL; cate a variety of wind tunnel models for different tests of multiple types
∙ Powder bed fusion—an additive manufacturing process in which of aero vehicles. The major of AM-based models for aircraft are listed in
thermal energy selectively fuses regions of a powder bed, commonly Table 1.
known as SLS.

2. Design procedures of AM-based models

1.3. AM-based models: a brief history
2.1. Requirements for models
In 1997, Springer et al. at NASA Marshall Space Flight Center first
introduced the AM for the design and fabrication of models for wind 2.1.1. General requirements for models
tunnel tests [25]. Several studies [10,25–27] have demonstrated the Wind tunnel models are reduced-scale alternatives to the prototypes
advantages of additive manufacturing techniques for building wind- of aerial vehicles used in wind tunnels. Wind tunnel models can be
tunnel models in less time and at a lower cost relative to traditional classified according to different classification criteria such types of
techniques. Dimensional tolerances and surface finish may be inferior, aerial vehicle prototypes, the structures of the aerial vehicle prototypes,
which in turn increases experimental uncertainty, but nevertheless the the types of data obtained, the similarity criteria met, the test speed
method is quite adequate for preliminary studies. A most advantageous [3,4], etc., as listed in Table 2.
aspect of the AM-based method is the ability to produce internal fea- The design and fabrication of wind tunnel models is a complex
tures, that would be difficult to implemented by traditional techniques process as it should meet similarity, measuring, strength and processing

Fig. 2. AM-based models used for various applications.

3
 W. Zhu

Table 1
Progresses in AM-based models.
Country Groups First author Year Air vehicles Model AM processes (materials)

Australia University of New South Wales Shun [19] 2012 airfoil force SL (resin)
Ahmed [50] 2013 full (UAV) force SLM (Carbonmide)
Brazil Institute of Advanced Studies Marcos [51] Rego [52] 2017 2018 entrance ramp (motor force FDM (Stratasys Fortus)
scramjet)
China Xi'an Jiaotong University, China Aerodynamics Research and Zhou [53] Zhou [54] 2007 2008 full (fighter, control face) force;pressure SL (resin)+ED (Nickel)
Development Center Zhu [55] 2011 wing (wing-box) dynamic SL (resin)
Yang [56] 2013 wing-body (AGARD-B) force SL (resin)
Zhu [57] 2014 full (fighter) force SL (resin)
Zhu [58] 2019 wing-body (F4) force SL (resin)
Iran K. N. Toosi University of Technology, Islamic Azad Daneshmand [59] Aghanajafi 2006 2006 wing-body-tail force SLS (glass reinforced nylon); SL
University [29] Daneshmand [60] 2007 (resin); FDM (ABS)
Nadooshan [61] 2007 wing-body (rocket) force FDM (PC); SL (resin)
Daneshmand [62] Daneshmand 2012 2012 wing-body (rocket) force FDM (ABS) & electroplating (Cr)
[63]
Daneshmand [64] 2013 wing-body (ARGARD-B) force FDM (ABS), 3DP
Italy Politecnico di Milano Gaspari [65] Gaspari [66] 2016 2018 wing (transport, control force SL (resin); SLS (PA)
surfaces)
Russia Russian Academy of Sciences Azarov [21] 2002 full (passenger/transport, force SL+casting (composition)
fighter)

4
Singapore Nanyang Technological University Teo [67] 2016 wing-body-tail (joined- force FDM (nylon)
wing)
US Alabama Univ., Huntsville Landrum [68] 1997 airfoil (2D) force SL (resin)
Cornelius [69] 1999 wing-tail (missile inlets) force SL (resin)
US NASA Marshall Space Flight Center Springer [26] Springer [25] 1997 1997 wing-body-tail (launch force SL (resin); FDM (ABS); SLS (glass
Springer [27] Springer [70] 1998 1998 vehicle) reinforced nylon)
US NASA Langley Buck [6] 2000 wing-body-tail (X-33 force SL+slip casting (ceramic, metal)
Phase I)
US Texas A&M University Strelec [20] 2002 airfoil force FDM (ABS)
US Hildebrand [71] Tyler [72] Tyler 2003 2005 wing-body (UAV X-45A) force; pressure; SL (resin)
Air Force Research Laboratory; Johns Hopkins University; University of Dayton Research [73] 2008 temperature;
Institute; Innovative Scientific Solutions, Inc.; Miami University; University of Dayton Tyler [74] Tyler [28] Gebbie [75] 2004 2004 full (striker tanker, force; pressure SLS (steel)
2005 blended wing-body)
Pankonien [76] Pankonien [77] 2017 2018 full (flying-wing aircraft, force SL (resin)
X-56A)
Reeder [78] 2007 full (transport E-8C) force SL (resin)
Black [79] 2018 airfoil static aeroelastic FDM (PLA)
US Honda Americas Fujino [80] 2003 wing-body (regional dynamic SL (resin)
transport)
US Embry Riddle Aeronautical University Traub [81] 2013 airfoil force FDM(ABS)
US Northrop Grumman Bartley-Cho [82] 2008 all-wing (dynamic model) dynamic SL (resin)
UK Imperial College London Heyes [22] 2004 airfoil force FDM (PC), SL (resin)
UK University of Bristol Woods [83] 2016 wing (morphing wing) smart wing FDM (ABS) & silicone skin
UK University of Bristol (UK) Ai [84] 2018 wing (morphing wing) smart wing SLS (polyamide)
Progress in Aerospace Sciences 110 (2019) 100541
W. Zhu Progress in Aerospace Sciences 110 (2019) 100541

Table 2 Model designing: The design of the wind tunnel model begins with
Types of wind tunnel models. the similarity design. In this step, the corresponding parameters of the
Classification criteria Wind tunnel models model are obtained from the geometry and structure parameters of the
aircraft prototype according to the similarity laws. Under the con-
Types of prototypes Aircraft, missile, rocket, etc. straints of all the requirements shown in Fig. 3., the detail structures of
Structures of prototypes Whole, half, wing, wing-body, etc.
the model can be obtained through tools such as CAD [28]. After ver-
Types of data Force, Pressure, dynamic properties, etc.
Laws of similarity Geometry, stiffness, mass, dynamic, etc.
ified by the step of design validation [86] (described in more detail
Speeds of testing Low speed, subsonic, transonic, supersonic, below), the CAD data of the model can be obtained.
hypersonic, etc. Data Processing: Data processing is used to generate a database
that drives the operation of AM machines. In general, common formats
for AM data include STL or AMF formulating the model parts in slicing
layers [23]. Therefore, it is necessary to convert CAD data into the
above data format. Fortunately, mainstream CAD design software now
provides a data processing interface. Supporting and Orientation are
necessary steps in the preparation of the AM process [87], which will be
described in detail below.
Model Forming: With the data, model parts are automatically
printed by the AM machine. For most AM processes, it is necessary to
remove the support and clean the residual raw materials to obtain the
green parts of the model. In order to improve material properties and
surface roughness, the model parts can be post-processed, such as post-
curing for SL parts [87], sand blasting for the surfaces of parts [85], etc.
Model assembling: Thanks to the powerful forming capability of
AM technology, the integrity of the model structure is improved, and
the number of model parts is greatly reduced [73]. This simplifies the
part assembling procedure. During the apparatus installing step, sensors
Fig. 3. Requirements for design and fabrication of models. and devices for data acquisition are to be installed on the model
[74,85]. After the assembled model is fixed to the wind tunnel support
(usually the balance [78]), the whole procedure for model development
requirements, as shown in Fig. 3. Among them, similarity requirements ends and the data is ready to be acquired from wind tunnel testing
are the most critical [12–14]. Wind tunnel models are designed ac- according to the Test programs and Operation instructions.
cording to the similarity laws [12] that determine the factors for
scaling-down from aerial vehicles (prototypes). These factors include 2.3. Model validation
geometric, mass, and dynamic similarity factors. The geometric simi-
larity means that the external contour of the model is consistent with To ensure the safety of the wind tunnel test, the designed model
the prototype to ensure the similarity of the flow, which is the basic must pass the strength calibration and the resonance clearance. To
criterion to be met by all models. The stiffness and mass similarity verify the design and manufacturing accuracy, ground stiffness cali-
determine the internal structure of the model so that it has the same bration and ground vibration calibration should be conducted on the
stiffness and mass distribution as the prototype, according to which assembled model.
static elastic models and free-flight models are designed. For dynamic
models, e.g. flutter models, the dynamic similarity should be met. 2.3.1. Strength validation
The purpose of the strength validation is to ensure that the model
2.1.2. Special requirements for AM-based models does not undergo structural damage during wind tunnel testing and to
Compared with traditional casting and machining processes, the prevent damage to the personnel or wind tunnel. The strength valida-
process limitations of AM are greatly reduced. It can help to simplify the tion of the model usually uses numerical simulation methods.
design and fabrication process of wind tunnel models. For AM-based Commonly used simulation methods include CFD and FEA [72,79].
wind tunnel models, however, special requirements should be con- Firstly, wind tunnel test conditions that will cause the maximum load
sidered, such as material selection and layered process characteristics. on the model should be determined. Worst-case analysis should be
Material selection: Limited to processing accuracy, cost, cycle, etc., conducted according to the wind tunnel testing scheme [57]. Secondly,
the main material used for the wind tunnel model is non-metallic ma- the load applied to the model is calculated based on the wind tunnel
terials, especially acrylic plastics [25]. When designing a wind tunnel test conditions, and this calculation relies mainly on CFD [57,58]. Fi-
model, the strength, modulus, water absorption, and heat resistance of nally, with the flow field from the CFD calculation used as the pressure
AM materials should be fully considered [10,72]. loading, the deformation and stress are calculated using the FEA
Layered process characteristics: For most AM processes, size method [57,74]. If the maximum stress of the model is less than the
constraints of AM equipment, supporting and removing of the parts, required stress of the material under the maximum load, the design is
cleaning of the interior material, fluctuations of material properties and checked by strength.
roughness caused by the forming direction should be considered [58].
2.3.2. Frequency clearance
2.2. Design procedure To avoid resonant damage of the model during testing, the first
eigen mode of the model-balance-support system is usually designed to
Based on AM technology, the whole design and fabrication proce- be higher than the peak frequency of the wind in tunnel due to the flow
dure of the model, including model designing, data processing, model fluctuation. The wind peak frequency varies with the wind tunnel
processing and model assembling, can be divided into two stages: adopted. The first eigen frequency of the system is determined by its
Virtual model and Physical model [85]. As shown in Fig. 4, the input stiffness, damping and mass properties. As the first two properties lar-
includes all the above-mentioned requirements, and the output includes gely depend on the support that cannot be changed, the model weight is
the final wind tunnel model and operation instruction. therefore the main designable factor. In general, less model weight

5
W. Zhu Progress in Aerospace Sciences 110 (2019) 100541

Fig. 5. Segmentation of AM-based models [28].

Fig. 4. Design procedure of models based on AM.

direction, it is difficult to ensure high precision for a complex model. On
the contrary, by segmenting a model into parts according to the contour
characteristics, the surface quality of the model can be improved.
leads to higher frequency and thus results in more safety [58]. In other However, if the number of model parts increases, the assembling error
words, due to its reduced weight the AM-based model has a great may rise. Therefore, unnecessary segmentation should be avoided and
prospect to improve the safety during wind tunnel testing [57]. the number of model parts should be kept as small as possible [74].
Generally, in order to ensure the aerodynamic contour of the model, the
2.3.3. Ground stiffness testing fabrication direction of the model should be along the axis of the air-
For models that need to meet stiffness similarity requirements (such craft, so that the segmenting surface of the model is perpendicular to
as static aeroelastic models, dynamically similar models, etc.), the as- the axis. The segmentation surface should be not too large, otherwise
sembled model needs to pass the ground stiffness testing to determine if the flatness of the plane will be difficult to guarantee. At the same time,
the stiffness distribution is as expected. The elastic axis location can in the model segmentation, the structural stress also needs to be con-
also be identified [88]. In a testing conducted by Wang et al. in the sidered [91]. An example of segmentation is shown in Fig. 5.
Aerodynamics Research and Development Center of China, the stiffness
distribution of the model was deduced from the model deformation that 3.1.2. Connection
was measured under specific loads at specific points using a contact-free 3.1.2.1. Gluing and screwing. There are usually two connection methods
measuring method. The Cartesian coordinates of the model on selected for segmented model parts. One is direct connection between parts,
points were measured and the deformation was calculated by com- such as gluing between AM plastics [71,74]; the other is connection by
paring the values before and after loading [58]. a third-party connector [57]. In order to improve the positioning
accuracy between the directly connected parts, some auxiliary
2.3.4. Ground vibration test positioning structures can be designed on the connection surfaces,
The purpose of the ground vibration test is to identify the dynamic such as groove features [19]. However, the direct connection still has
characteristics of the model, such as eigenfrequencies, eigenmodes, problems such as insufficient connection strength and low connection
damping properties, etc., to check the fabrication accuracy of dynamic accuracy. Therefore, indirect connecting through the third-party
models [82,88,89]. In a typical test, the model is first mounted on a connector is a better choice. The parts are connected to the third-
special fixture, and then the response in the time domain is acquired via party connector made of machined metals, and the gaps between the
sensors under excitation by hammers or exciters. After the data is parts are filled with glue or foam to ensure the integrity of the
processed with FFT (Fast Fourier Transformation), the eigenfrequencies aerodynamic shape [80]. It is worth mentioning that the third-party
and eigenmodes of the model can be obtained. Different dynamic connector is not necessarily a separate component, and some functional
parameters need to be obtained through different excitation methods components with connection function can also be used. For example,
[90]. Non-contact vibration testing methods are used to reduce sensor the metal spar in the spar-shell structure can act as the third-party
interference and improve test efficiency [87]. In our own research, a connector of the AM plastic shells. The overall assembly accuracy of the
laser-vibration-measuring apparatus was employed to test the dynamic model depends mainly on the assembly accuracy between the AM parts
behaviors of the model. Each measuring point was scanned under the and the third-party connector, especially on the fabrication precision of
excitation of an impact-hammer [55]. the AM positioning/mounting holes. The precision of the AM holes can
be improved by presetting a certain size compensation and optimizing
3. Fabrication processing of AM-based models the corresponding process parameters (orientation, etc.). However,
holes with curved surfaces are still difficult to achieve. At the same
3.1. Segmentation and connection time, the model usually needs to be disassembled many times, and the
plastic holes will lose the positioning accuracy due to the inevitable
3.1.1. Segmentation wear during the disassembly and assembly process. To this end, Zhu
At present, the maximum building size of the existing AM is gen- et al. [57] proposed to install a metal bushing in the resin mounting
erally less than 1 m [24]. If the size of the model part exceeds the hole. On the one hand, the metal bushing avoids the protection of the
building range of AM machines, the model cannot be fabricated in one plastic from directly participating in the assembly; on the other hand,
cycle by AM. In this case, the model needs to be segmented into several the metal bushing can adopt an external positioning structure to avoid
smaller parts, and then bonded or assembled into the whole model the precision degradation caused by the direct positioning of the resin
[28,71,74]. If the whole model is fabricated by AM via a single hole.

6
W. Zhu Progress in Aerospace Sciences 110 (2019) 100541

3.1.2.2. Tenoning. For models that need to simulate structural

properties such as stiffness distribution, mass distribution, and force
transfer, the above gluing or screwing methods might have an adverse
effect on structural properties. Thanks to the AM forming capability, a
new tenoning connection was proposed (as shown Fig. 6) by Pankonien
et al. at the US Air Force Research Laboratory [76]. The tenoning
connection utilized complex interleaving geometries across the
connection surface. These geometries propagate only to the existing
filled volume within the AM-built parts, which then preserves the total
volume of the existing material and thus its mass and stiffness. This
method has two advantages over the above-mentioned gluing and
screwing. The distributed nature of the tenoning connection spreads the
fastening over a larger area, thus reducing the effects of localized stress
concentrations on the dynamic response of the structure.

3.2. Roughness

Similar to the effect of the aircraft surface properties on the aero-

dynamic characteristics, the surface properties of the model also has an
impact on the aerodynamic data in the wind tunnel testing [92]. In
general, the AM models do not have smooth surface because of the
surface roughness caused by the layer-by-layer forming principle of AM.
Therefore, it is necessary to evaluate the influence of the surface to-
pography on aerodynamic characteristics, and it is necessary to study
methods to improve the surface properties of the model.
Fig. 7. Roughness of models smoothed by different processes [81].
3.2.1. Effects of surface properties on the aerodynamic characteristics
The effects of surface finish and grit on the aerodynamic char- roughness criteria, which is based on two-dimensional boundary layer
acteristics of the models were investigated in many studies [29,61,93] flow.
[81]. It can be concluded that surface finish does have an effect on the
aerodynamic characteristics, but the application of grit has little effect
on the aerodynamic characteristics except on the axial force and its 3.2.2. Smoothing of AM model surfaces
derivative coefficients [60,93,94]. Gebbie et al. at the US Air Force A study by Aghanajafi et al. [93] investigated surface-finish effects
Institute of Technology used SLS to fabricate a stainless-steel model of a through variation in the layer thickness of models produced using SL.
strike tanker in the blended-wing-body configuration [75]. The model Their results indicated that greater layer thickness impacted the axial
surface was polished and coated with PSP painting to the sub-micron force (and consequently drag) to the greatest extent. Post-processing by
roughness. The effect of the model roughness was investigated by grinding and painting is necessary in order to achieve a satisfactory
comparison of lift curves for the model with the above-mentioned surface quality. The painting was carried out with spray paint, with the
coverages. It can be concluded that the roughness affected the forces application of several thin layers. This process was carried out twice
acting on the model. Schairer et al. at the NASA Ames Research Center [85]. Finishing was accomplished using paint and sanding. The airfoil
investigated the sensitivity of the roughness effect to Reynolds number finish was manipulated using both sanding and painting. The surface
by comparison of forces and pressures for a wing model with and morphology is depicted in Fig. 7. The results suggest that for this airfoil
without PSP painting [95]. It was found that at high Reynolds numbers, and test conditions, the surface preparation had a minimal impact on
the roughness due to the painting changed the pressure distributions. the measured loads [81]. The surface roughness for the FDM model and
However, at low Reynolds numbers the roughness effect was quite the model with electroplating coating was 16 (Ra) and 0.832 μm (Ra)
small. The results were consistent with the classical admissible [62].

Fig. 6. Tenon connection of model segmentation [76].

7
W. Zhu Progress in Aerospace Sciences 110 (2019) 100541

3.3. Model support and orientation

For most AM processes, internal supports are the necessarily aux-

iliary structures to prevent the upper components of the models from
collapsing due to gravitation, but they are very difficult to clean later to
ensure the similarity of the internal structures of the cavity-like model
[55]. As demonstrated in Fig. 8, with the increase of the angle, less
internal supports are needed as more force to support the upper com-
ponents is provided by the cured structures (indicated as Fc) instead of
internal supports or buoyancy (indicated as Fs). Though larger angle
seems favorable to avoid internal supports, the resulting considerable
increase of the building time has to be taken into account. Therefore, Fig. 9. SL Models before and after electroplating [53,54].
30°was selected as the angle of the building orientation to avoid any
internal supports while minimizing the building time. This question is that the strengthening method by using electrodeposited nickel coating
important in that additional support structures require an increased use is feasible to be used in a subsonic and a high subsonic (Ma < 0.9) wind
of material. In addition, forgoing a support structure can avoid the te- tunnel with larger AoA.
dious washing of the supports in an alkaline bath in post-processing or
at least considerably reduce the manufacturing time. Thus, one can
4. Applications of AM-based models
already recognize in the pre-processing that from an angle of 40° be-
tween the basis and the braces no support structures are created by the
4.1. Models for force measurements
software. A sample part is additionally created for demonstration which
explains this circumstance [96].
The wind tunnel testing for force measurements is the most basic
test that must be done for all types of aircraft during R&D. The aero-
3.4. Model strength dynamic data, lift, drag, pitching moment etc., should be acquired for
varying Ma numbers, angles of attack, etc. In some tests, the aero-
The materials that are currently widely used in AM models are dynamic characteristics of the control surfaces should also be measured.
mainly non-metallic, especially plastic materials such as acrylic resins The wind tunnel model for force measurement requires complex in-
used by SL [56,60] and Polyjet [97], polycarbonates [61], ABS (Acry- ternal structures to install different measuring sensors, and must meet
lonitrile Butadiene Styrene) [63,83] and Nylon [62] used by FDM, the requirements of high manufacturing precision, short cycle, and low
polyamides [65,84] used by SLS, etc. For high-speed or high angle of cost.
attack tests, which may cause large loads, the strength of plastic ma-
terials is insufficient [10,27]. Plastic models have limited functionality
4.1.1. Models for subsonic/transonic tests
owing to low strength and stiffness, poor creep performance, and en-
Tyler et al. at the US Air Force Research Laboratory designed,
vironmental instability (moisture and temperature). Electroplating is
analyzed and fabricated a wind tunnel model with a lambda wing-body
used to strengthen the materials properties of AM plastics
configuration based on SLS [71,72]. The outer surface was used as a
[53,54,62,63]. For thin model components, e.g. wings, flaps, ailerons
starting point for the model while the internals of the model, including
etc., a new strengthening method by using electrodeposited nickel
static pressure taps and the wind tunnel balance, were drawn prior to
coating was proposed by Zhou et al. at Xi'an Jiaotong University of
the model building process [28,72]. An Iranian joint group, formed by
China [53,54]. The mechanical test results of nickel-coated Stereo-
Daneshmand et al. at Islamic Azad University and Aghanajafi et al. at K.
lithography (SL) samples show that coatings of electrodeposited nickel
N. Toosi University of Technology, tried many AM processes to evaluate
on SL prototypes result in increases in Young's modulus, UTS, flexural
their feasibility for the fabrication of models, e.g. SLS [29,59], FDM
modulus, and strength. Through the comparative analysis of aero-
[61], three-dimensional printing [64,98–100], SL [61,93], etc. It was
dynamic characteristics between different coated SL models as shown in
undertaken to determine the suitability of the model constructed with
Fig. 9 and a metal model, it is found that the deformation of the
SLS method in subsonic and transonic wind tunnel testing (Ma 0.3 to
composite model with a nickel-coated wing is less than with a bare SL
1.2). Fused Deposition Modeling (FDM) was used to show its ability to
wing, and the aerodynamic characteristics of a nickel coated wing
make components for wind tunnel models in a timely and cost effective
model are closer to a metal model than a SL wing model. It is proved
manner [61]. The manufactured models using a three-dimensional
printing method can be used for primary tests, being less expensive and
requiring significantly less time to build [64,98–100]. For models fab-
ricated by SL, the layer thickness has an increased effect on the aero-
dynamic characteristics when the Mach number is decreased and has
the greatest effect on the axial force and its derivative coefficients
[61,93]. Junk et al. at University of Applied Sciences Offenburg of
Germany manufactured the UAV parts by means of the binder jetting
process [85]. Bykerk et al. at University of Sydney has used these
printers in the construction of hypersonic aircraft models for testing in a
low speed wind tunnel. Larger models were printed in ABS (Acryloni-
trile Butadiene Styrene) parts, then assembled and post-processed to
ensure continuity between the original design and the model. Techni-
ques include sanding, gap-filling, sanding again, resin-coating and
painting [101]. Polyamide (PA) 2200 and nylon were used to manu-
facture the joined-wings by Teo et al. at Nanyang Technological Uni-
versity, Singapore. Nylon was chosen after material characteristic
strength, surface finishing, cost and time were considered [67,102].
Fig. 8. Relationship between building orientations and internal supports [55]. Hammond at Airbus UK used SL resin to made parts of the bathtub and

8
W. Zhu Progress in Aerospace Sciences 110 (2019) 100541

Fig. 10. AM-based models with fixed angle control surfaces [71,72].

wing-to-body fairing design for an A3XX transonic model [103]. Udroiu technique not only can help to efficiently build flexible models but also
et al. showed that Polyjet technology can be effectively applied for provide a new testing scheme for the conceptual design of aircraft.
fabricating test models with exceptionally high quality, accuracy and
speed [97]. 4.1.4. Models with control surfaces
In the development of aircraft, many models need to be tested with
4.1.2. Models for hypersonic tests control surfaces as shown in Fig. 10 [71,72].
Buck at the NASA Langley Research Center reviewed the accurate
model casting techniques for fabrication of benchmark metal and 4.1.4.1. Varying control surfaces via different models. Reeder et al. at the
ceramic test models using a combination of AM-based patterns and US Air Force Institute of Technology conducted wind tunnel
investment casting [6]. The technique is a major step forward in the measurements on a variety of E-8C configurations built by SL [78].
area of missile models for high-speed tests [69,104]. Marcos et al. at The control surface and the main structure of the models were designed
Fundação de Ciência of Brazil and Rego et al. at Instituto de Estudos as an integrated structure. The influence of control surface deflection on
Avançados of Brazil evaluated the feasibility of 3D printed models used the aerodynamic characteristics was investigated through a series of
for supersonic tests [51,52,105]. The experiments showed that 3D models.
printed models can be used in hypersonic shock tunnels as long as they
were reinforced to resist the severe structural stresses caused by the 4.1.4.2. Varying control surfaces via connectors. For the AM-made plastic
high loads in hypersonic flow. Danehy et al. at the NASA Langley Re- adjustable control surfaces, Zhu et al. adopted a more complicated
search Center used SL to fabricate reentry models for use in hypersonic connection mechanism [57]. As shown in Fig. 11, the control surfaces
wind tunnel tests. These models were produced in one day or less, were fixed onto the load-bearing bases by bolts with specialized V-
which is a significant time savings compared to the manufacture of shaped brackets as the intermediary. The leading-edge flaps, trailing-
ceramic or metal models. The models were tested at the NASA Langley edge flaps, and ailerons were supported by the metal cores on the wing,
Research Center 31-Inch Mach 10 Air Tunnel. Most of the models did whereas the rudder was fixed onto the resin tail itself. Brackets
not survive repeated tests in the tunnel, and several failure modes of the (connectors) with specialized angles between the two arms according
models were identified [106,107]. to the deflection angles of the control surfaces were changeable and
acted as the deflection adjusters of the control surfaces. Again, all the
4.1.3. All plastic models for transonic tests connection surfaces on the resin parts were covered by metal bushes to
Rigid models made of materials with high elastic moduli are widely avoid abrasion.
used for high-speed wind tunnel testing. In general, models fabricated
only by plastics cannot be used for the purpose due to their large de- 4.2. Models for surface-pressure measurements
formation caused by their lower strength and modulus. Zhu et al. de-
veloped a method to design and fabricate all-plastic models for 4.2.1. Models based on direct pressure measurement
transonic tests [58]. The model was designed with pre-deformed con- The main purpose of the wind tunnel tests for surface-pressure
tours which can deform to the target contour under aerodynamic loads.
The key is to obtain the aerodynamic contour of the model for fabri-
cation, the designed contour, that can transform into the target contour
under aerodynamic loads in a particular operation state. As mentioned
above, the AM-built plastic models are flexible, so the original contour
for fabrication should be pre-deformed downward to a suitable position
that can deform upward back to the target contour under aerodynamic
loads, which makes it is possible to acquire data for the deformed
(target) contour without deformation corrections. An CFD/CSD cou-
pling method was used to compute the pre-deformation, based on
which the models were designed. The models were fabricated by AM
and tested in a wind tunnel. The plastic models were undamaged at
Mach numbers up to 0.85. Furthermore, the cost and time reduction,
made by the introduction of AM due to its fabrication flexibility, can
potentially improve the R&D efficiency of aircraft. Thanks to the low
modulus of its materials and high efficiency of its process, the AM Fig. 11. Structures of the deflection adjustment mechanism [57].

9
W. Zhu Progress in Aerospace Sciences 110 (2019) 100541

Fig. 12. Surface pressure SL model with taps and passages [53,54].

measurements is to provide the original data of the aerodynamic load

distribution for the strength calculation of the aircraft. It provides a
basis for studying the aerodynamic performance of the aircraft and its
components, and studying the flow characteristics around the model.
The position of the minimum pressure points on the wing, the separa-
tion characteristics of the airflow, and the lift, differential pressures,
and position of the pressure center acting on the model can be de-
termined. The pressure measuring model has pressure measuring holes
along the normal direction of the surface at positions where the pres-
sure measurement is required. The holes are connected to a measuring Fig. 13. CFD (left) and PSP (right) results for a blended wing-body configura-
device such as a pressure gauge or a pressure sensor, through the tion [75].
pressure measuring tubes. Compared to conventional pressure mea-
suring models, AM has advantages in accommodating complex internal
structures for pressure measurement models with internal bores. 4.3. Aeroelastic models
Therefore, it can be expected to reduce machining costs and increase
machining efficiency. Zhou et al. [53,54] studied the effects of the hole 4.3.1. Static aeroelastic models
forming defects, such as the shape of the hole edge, the inclination of The determination of static aeroelastic effects requires the use of
the hole axis, the protrusion (or depression) near the hole, the scratch, static aeroelastic wind tunnel models [2,109]. To quantitatively re-
etc., on the pressure measurement (see Fig. 12). present the real aircraft, the static aeroelastic model should meet si-
Modern 3D printing techniques are ideally suited for the manu- milarity requirements, such as geometric and stiffness similarity [4].
facturing of AM wind tunnel models. As shown by Shun and Ahmed at Based on additive manufacturing technology, a static aeroelastic
University of New South Wales, Australia, their accuracy yields reliable model with an integrated resin shell was designed, calibrated, and
data [19]. Torres et al. used 3D printing for the manufacturing of a tested by Wang et al. and Zhu et al. [89,110]. The new structure of the
wind tunnel model of a NACA 65 (2)-415 airfoil, representative of the model can help improve wind tunnel test efficiency. Bartley-Cho and
cross section of an Inlet Guide Vane (IGV) in an industrial gas com- Henderson at Northrop Grumman Corporation used SL to directly
pressor [108]. Tyler et al. at the US Air Force Research Laboratory manufacture pods of an aeroelastic model [82]. Several changes and
made a model with deflecting control surfaces using SL, which had 64 enhancements were added to the model to fit the test needs. These
pressure taps integral to the non-metallic parts and two thermocouples changes/enhancements included replacement of all SL parts as shown
[74]. Heyes and Smith at Imperial College London also used SL to in Fig. 14.
fabricate a surface-pressure model, that had a chord of 70 mm and
pressure taps of 0.35 mm in diameter [22]. 4.3.2. Dynamic models
Aeroelastic models are often used in wind tunnel tests to investigate
4.2.2. Models based on PSP (pressure sensitive paint) measurement flutter and other aeroelastic characteristics or to verify the aeroelastic
Gebbie et al. at the US Air Force Institute of Technology manu- numerical simulation of aircraft or their components [90,111]. To map
factured a wind tunnel model of a blended wing body configuration by the data of the tests to the actual flights, they should be designed and
selective laser sintering (SLS) [75]. Subsequently, they applied PSP to fabricated according to pertinent similarity laws which determine the
the model to enable global pressure measurements for comparison to required tunnel density and velocity of the external contour, stiffness
CFD computations. Force and moment data were taken along with the and mass distribution of the models, etc. The most important similarity
PSP. However, dramatic differences were seen in the measured force laws include: geometric, stiffness and mass similarity [112].
and moment data before and after paint had been applied to the model. Fujino et al. at Honda R&D Americas, Inc. evaluated the basic flutter
The test conditions between the unpainted and painted model test were characteristics of an over-the-wing engine mount configuration. The 1/
constant with the exception of the application of the paint to the model. 4-scale model employed spar-strip construction [80]. The aluminum
As shown in Fig. 13, both the PSP data and the CFD results indicated spar plus flange section provided the required stiffness distribution in
that the primary mechanism of lift transitioned from leading-edge bending and torsion. The strips were made of SL resin and lead and
suction along the wing to vortex lift near the wing-body junction as tungsten weights were installed in each strip to yield the required mass,
angle of attack increased. The data suggests that, for the range of center of gravity, and moment of inertia properties. The gaps between
conditions tested, submicron roughness has a pronounced effect on the the strips were aerodynamically sealed with sponge rubber as shown in
transition to the vortex lift mechanism for increasing angle of attack. Fig. 15.

10
W. Zhu Progress in Aerospace Sciences 110 (2019) 100541

Fig. 14. The static aeroelastic model as installed in the wind tunnel [82].

Table 3
Economic comparison of AM technology with the traditional technology
(benchmark).
Models AM Benchmark Cost (AM/ Time (AM/
NC) NC)

Force models [29] SLS, FDM, SL NC(Al) 1/5 1/5-1/10

Force models [61] FDM NC(Steel) 1/2 1/2-1/4
Force models [62,63] FDM NC(Steel) 1/3-1/2 1/3-1/5
Force models [58] SL NC(Steel) N.A. 1/5-1/6
Hypersonic models SL+Deposition NC 1/8-1/10 N.A.
[69,104]
Dynamic models [55] SL NC(Steel) 1/7 1/10

5. AM applications to new vehicles

5.1. Unmanned aerial vehicles (UAVs)

Fig. 15. The dynamic model used for flutter test [80].

The development and use of unmanned aerial vehicles (UAVs) have

Pankonien et al. at the US Air Force Research Laboratory proposed a greatly increased in recent years for both commercial and military
new fabrication method for low-speed wind tunnel flutter models by applications. AM offers several advantages compared to conventional
exploring the role of AM to precisely control geometry and localized manufacturing methods [50,96,114]. Therefore, many groups have
stiffness. Thereby a new class of flutter models with increased geo- tried to fabricate UAVs based on AM as summarized by Goh et al. at
metric accuracy and stiffness control was enabled [76]. Raskin at Tufts Nanyang Technological University, Singapore in Fig. 16 [114].
University manufactured an AM multi-material wind tunnel model to The first example of a UAV that was manufactured with the aid of
investigate the possibility of using PolyJet AM to create a model that SLS is“SULSA” which was developed by Marks at University of
exhibits aeroelastic behavior [13]. A method to design and fabricate Southampton, UK [115], followed by the VAST AUAV based on FDM by
dynamic wing models for wind tunnel tests was proposed by Zhu et al. Stern and Cohen at Massachusetts Institute of Technology [116]. AM
based on Stereolithography (SL) [55]. This method can ensure the offers advantages which are of significance specifically for the manu-
structural similarity of both external and internal structures between facturing of UAVs. Various plastics can be introduced alongside metals
models and prototypes. The SL-based method could help to improve the to enable a light construction. Also, the layer-by-layer manufacturing
prediction accuracy of models in two aspects: ensuring structural si- allows for the application of complex forms with free-form surfaces for
milarity between prototypes and models, while reducing the total mass the wings and tails. Furthermore, complex internal features, such as
of the balance-weights significantly. joists, stringers and ribs in the airframe, can also be easily fabricated.
Finally, additional geometric details, such as, for example, supports for
sensors or openings for cables, can easily be integrated into the com-
4.4. Economic comparison ponents of a UAV without any additional treatments.

Economic considerations such as cost and time are the main moti- 5.2. Morphing wings and adaptive structures
vations for introducing AM to design and fabricate models. Using tra-
ditional machining methods, the cost of a typical wind tunnel model is The 3D printing technology has been identified as making it possible
approximately $100 k [69,104,113] requiring approximately half a to manufacture morphing devices in a single piece with the required
year [56,89]. Many studies were done on economic comparisons of accuracy in thickness distribution (of the order of magnitude of
models. As can be seen from Table 3, the cost and time of AM-based 0.5 mm). The best 3D printing techniques and the most suitable mate-
models are greatly reduced compared to the traditional method. For rials include the SL, using Somos GP Plus 14122, and the SLS, using PA
force models, the cost and time of the AM-based models are reduced by 2200 [65,66]. Ai et al. at University of Bristol, UK adopted SLS to
more than 1/2. For hypersonic models, the cost is reduced to about 1/ manufacture variable stiffness morphing trailing edge flaps [84].
10. For dynamic models, the cost and time are reduced to approxi- Woods et al. at University of Bristol used an FDM printer to manu-
mately 1/7 and 1/10 respectively. In general, the higher the design facture a single monolithic model that can morph [83]. To illustrate the
requirements and complexity of the model, the more prominent the potential for 3D-printing to realize designs with non-conventional
superiority of AM technology. topologies to address complex aeroelastic requirements, Pankonien

11
W. Zhu Progress in Aerospace Sciences 110 (2019) 100541

Fig. 16. Timeline representing AM fabricated UAVs [114].

Fig. 17. FDM-fabricated SMA wings [20].

et al. at the US Air Force Research Laboratory used a multi-material 3D At present, the dimensional precision of the main AM processes is
printing method to fabricate an integral trailing edge control surface about ± 0.1 mm. Since AM is based on the layer-by-layer principle,
that was adaptive and cannot be implemented with traditional techni- causing step effects AM-based models with curved aerodynamic con-
ques [77]. tours do not meet the requirements. For these problems, a feasible so-
To create an adaptive wing whose airfoil can achieve the potential lution is to combine AM and NC machining. AM-based models are
large-scale deflections necessary to maintain high efficiency as flow mainly made of non-metallic materials, especially plastics. The plastic
conditions change. Strelec and Lagoudas at Texas A&M University used models cannot be used for small structures with large bearing forces
the FDM method to manufacture an adaptive wing model with a shape (such as control surfaces) in high-speed tests. In addition, the plastic is
memory alloy spar inside as shown in Fig. 17 [20]. The study showed subject to aging, which makes the plastic model unsuitable for long-
the benefits of using the FDM generated plastic skin so that the thick- term tests. Also, AM processing causes fluctuations in material prop-
ness can be varied throughout the model, whereas the steel skin has a erties, which may produce uncertainties in the wind tunnel test results.
constant thickness. By changing the thickness of the skin, precise areas In the future, the processing precision and surface roughness will be
on the wing can be tailored to be stiffer than the rest of the wing. This improved, and the AM-based models will be used in more wind tunnel
variable skin thickness could be used in future efforts as a design op- tests. The development of cold spray additive manufacturing tech-
timization tool. The AM method can also help to decrease the com- nology will facilitate the efficient fabrication of metal models.
plexity of the model. This decreased complexity of rib manufacturing Therefore, it is expected that metal models will become more widely
eliminates the need for a shaped rib surface. used in high speed wind tunnel tests. Also, AM will allow the fabrica-
tion of bearing structures of models and of sensors and circuits. This
6. Conclusions and future trends will help to improve the model's ability to acquire aerodynamic data for
the development of new air vehicles, such as flexible wings based on
The design and fabrication of wind tunnel models play an important smart materials.
role in the development of air vehicles. After years of development, AM-
based models have been widely used for force and pressure as well as Acknowledgements
for aeroelastic measurements Studies have shown that the introduction
of AM can reduce the difficulty of model processing, thus greatly im- This work was supported by the National Natural Science
proving the processing time and cost of the model. Furthermore, due to Foundation of China, P. R. China [Grant No. 51505457], the National
its multi-material integrated processing capability, AM is well suited for Science and Technology Major Project, P. R. China[Grant No. 2017-VII-
directly fabricating of air vehicles and not just models. 0008-0101], the Key Research and Development Program of Shaanxi

12
W. Zhu Progress in Aerospace Sciences 110 (2019) 100541

Province, P. R. China [Grant No. 2018ZDXM-GY-059], the Open Fund [25] A. Springer, K. Cooper, Comparing the aerodynamic characteristics of wind tunnel
of State Key Laboratory for Manufacturing Systems Engineering, P. R. models produced by rapid prototyping and conventional methods, 15th Applied
Aerodynamics Conference, AIAA, Atlanta, GA, US, 1997AIAA 97-2222.
China [Grant No. SKLMS2016013], the Fundamental Research Funds [26] A. Springer, K. Cooper, F. R, Application of rapid prototyping models to transonic
for the Central Universities, P. R. China, and the Youth Innovation wind-tunnel testing, 35th AIAA Aerospace Sciences Meeting & Exhibit, AIAA,
Team of Shaanxi Universities, P. R. China. The author would like to Reno, NV, 1997AIAA 97-0988.
[27] A. Springer, Evaluating aerodynamic characteristics of wind-tunnel models pro-
thank Prof. Dichen Li at Xi'an Jiaotong University, P. R. China, Dr. duced by rapid prototyping methods, J. Spacecr. Rocket. 35 (1998) 755–759.
Zhengyu Zhang and Dr. Yan Sun at State Key Laboratory of [28] C. Tyler, M.F. Reeder, W. Braisted, J. Higgins, L.D. Gebbie, Rapid Technology
Aerodynamics, P. R. China for their constructive suggestions. The au- Focused Experimental and Computational Aerodynamic Investigation of a Strike
Tanker. USAF Development Test and Evaluation Summit, AIAA, 2004 AIAA 2004-
thor would like to thank Prof. Maximilian Platzer at US Naval 6870.
Postgraduate School for his great efforts to improve the paper writing. [29] S. Aghanajafi, R. Adelnia, S. Daneshmand, Production of wind tunnel testing
models with use of rapid prototyping methods, WSEAS Trans. Circuits Syst. 5
(2006) 555–561.
Appendix A. Supplementary data
[30] H. Seifi, A.R. Javan, S. Xu, Y. Zhao, Y.M. Xie, Design optimization and additive
manufacturing of nodes in gridshell structures, Eng. Struct. 160 (2018) 161–170.
Supplementary data to this article can be found online at https:// [31] B. Linke, I. Garretson, F. Jan, M. Hafez, Integrated design, manufacturing and
doi.org/10.1016/j.paerosci.2019.05.001. analysis of airfoil and nozzle shapes in an undergraduate course, in: L. Wang,
L. Fratini, A.J. Shih (Eds.), 45th Sme North American Manufacturing Research
Conference, Elsevier Science Bv, Amsterdam, 2017, pp. 1077–1086.
References [32] E. Black, S.C. Stoll, Residential 3D model wind tunnel pressure testing, in:
A.E.D. DeLeon, C.J. Rossell, A.M. Dolhan, M.J. Drerup, J.B. Kardon, M.P. Lester
(Eds.), Seventh Congress on Forensic Engineering, American Society of Civil
[1] V.Y. Neiland, Review of TsAGI Wind Tunnels, (1992). Engineers, Miami, Florida, 2015, pp. 449–467 ISI BF4JP.
[2] J. Lerner, U. Boldfes, Wind Tunnels and Experimental Fluid Dynamics Research, [33] J. Giannatsis, V. Dedoussis, D. Karalekas, Architectural scale modelling using
InTech, 2011. stereolithography, Rapid Prototyp. J. 8 (2002) 200–207.
[3] J.D. Pereira, Wind Tunnels: Aerodynamics, Models and Experiments, Nova Science [34] S. Zhou, X.G. Hua, Z.Q. Chen, W. Chen, Experimental investigation of correction
Publishers, Inc., 2010. factor for VIV amplitude of flexible bridges from an aeroelastic model and its 1:1
[4] J.B. Barlow, W.H. Rae Jr., AP. Low Speed Wind Tunnel Testing, John Wiley & section model, Eng. Struct. 141 (2017) 263–271.
Sons, Inc., 1999. [35] P. Bere, C. Neamtu, Design and manufacturing methodology for F1 nose car, 2014,
[5] K. Fujii, Progress and future prospects of CFD in aerospace—wind tunnel and in: D. Popescu (Ed.), International Conference on Production Research, Technical
beyond, Prog. Aero. Sci. 41 (2005) 455–470. University of Cluj-Napoca, Cluj Napoca, ROMANIA, 2014, pp. 21–26 ISI BB8DD.
[6] G.M. Buck, Rapid Model Fabrication and Testing for Aerospace Vehicles. 38th [36] A. Kochan, Rapid prototyping helps Renault F1 team UK improve championship
Aerospace Sciences Meeting & Exhibit, AIAA, Reno, Nevada, 2000 AIAA 2000- prospects, Assemb. Autom. 23 (2003) 336–339.
0826. [37] R. Edwards, D. Forsman, A rapid prototyping application in wind tunnel testing a
[7] R.B. Mora, Unmanned Aircraft Wind Tunnel Testing. Advanced UAV student project, 2002 Annual Conference, ASEE Conferences, Montreal, Canada,
Aerodynamics, Flight Stability and Control, John Wiley & Sons, Ltd, 2017, pp. 20027.98.91 - 97.98.96..
181–199. [38] C.S.H. Lin, C.K. Chua, J.H. Yeo, Design of high performance badminton shuttle-
[8] M. Melanson, M. Chang, W. Baker Ii, Wind tunnel testing's future: a vision of the cocks: virtual and rapid prototyping approach, Virtual Phys. Prototyp. 8 (2013)
next generation of wind tunnel test requirements and facilities, 48th AIAA 165–171.
Aerospace Sciences Meeting Including the New Horizons Forum and Aerospace [39] F. Guerrero-Villar, E. Torres-Jimenez, R. Dorado-Vicente, J.I. Jimenez-Gonzalez,
Exposition, AIAA, Orlando, Florida, 2010AIAA 2010-142. Development of vertical wind turbines via FDM prototypes, Procedia Eng. 132
[9] M.L. Skelley, T.F. Langham, W.L. Peters, Integrated Test and Evaluation for the (2015) 78–85.
21st Century. USAF Developmental Test and Evaluation Summit, AIAA, Woodland [40] K. Bassett, R. Carriveau, D.S.K. Ting, 3D printed wind turbines part 1: design
Hills, California, 2004 AIAA 2004-6873. considerations and rapid manufacture potential, Sustain. Energy Technol. Assess.
[10] R.N. Chuk, V.J. Thomson, A comparison of rapid prototyping techniques used for 11 (2015) 186–193.
wind tunnel model fabrication, Rapid Prototyp. J. 4 (1998) 185–196. [41] H. Yu, N.A. Ahmed, T.G. Flynn, Manufacture and material considerations in wind
[11] T.W. Butler, Model Design and Fabrication Improvements for Wind Tunnel tunnel experimentation, in: P.S. Wei (Ed.), Machine Design and Manufacturing
Testing. US Air Force T&E Days, AIAA, Nashville, TN, United states, 2005, pp. Engineering III, Trans Tech Publications Ltd, Stafa-Zurich, 2014, pp. 694–699.
363–369. AIAA 2005-7639. [42] H. Wu, D. Li, N. Guo, Fabrication of integral ceramic mold for investment casting
[12] C.P. Coutinho, A.J. Baptista, J.D. Rodrigues, Reduced scale models based on si- of hollow turbine blade based on stereolithography, Rapid Prototyp. J. 15 (2009)
militude theory: a review up to 2015, Eng. Struct. 119 (2016) 81–94. 232–237.
[13] B. Raskin, Aeroelastic Characteristics of a Rapid Prototype Multi-Material Wind [43] T. Zhang, H.C. Yang, C.M. Miyamoto, 3D printing: a cost effective and timely
Tunnel Model of a Mechanically Deployable Aerodynamic Decelerator, Tufts approach to manufacturing of low-thrust engines, 50th AIAA/ASME/SAE/ASEE
University, 2016 ProQuest 10118393. Joint Propulsion Conference, AIAA, Cleveland, OH, 2014AIAA 2014-3502.
[14] A.J. Dutson, Functional Prototyping through Advanced Similitude Techniques, [44] P. Rochus, J.-Y. Plesseria, M.V. Elsen, J.-P. Kruth, R. Carrus, T. Dormal, New ap-
The University of Texas at Austin, 2002 UMI 3110601. plications of rapid prototyping and rapid manufacturing (RP/RM) technologies for
[15] J. Florance, J. Heeg, C. Spain, T. Ivanco, C. Wieseman, P. Lively, Variable Stiffness space instrumentation, 57th International Astronautical Congress, International
Spar Wind-Tunnel Model Development and Testing. 45th AIAA/ASME/ASCE/ Astronautical Congress, 2006IAC 06-C2.8.05.
AHS/ASC Structures, Structural Dynamics & Materials Conference, AIAA, 2004 [45] R.O. Crumpacker, Design of a modular vortex tube engine air particle separator for
AIAA 2004-1588. the MQ-1 UAV utilizing additive manufacturing technologies, 55th AIAA
[16] C. Gionet, F. Yandrasts, Developments in Model Design and Manufacturing Aerospace Sciences Meeting, AIAA, 2017AIAA 2017-1235.
Techniques. Royal Aeronautical Society 1997 European Forum, The Royal [46] K. Takato, S. Shirayama, Development of a 3D-printed device evaluating the
Aeronautical Society, Cambridge, 1997 RAS 1997. aerodynamic performance of rotary wings, J. Adv. Mech. Des. Syst. Manuf. 12
[17] C. Ciancarelli, M. Melanson, W. Baker, Wind tunnel test technologies for aircraft (2018) 11.
cycle time reduction, 37th AIAA Aerospace Sciences Meeting and Exhibit, AIAA, [47] J. Quincieu, C. Robinson, B. Stucker, T. Mosher, Case study: selective laser sin-
Reno, Nevada, 1999AIAA 99-0177. tering of the USUSat II small satellite structure, Assemb. Autom. 25 (2005)
[18] J. Kegelman, Recent cycle time reduction at Langley research center, 37th 267–272.
Aerospace Sciences Meeting and Exhibit, AIAA, Reno, NV, 1999AIAA 99-0178. [48] L. Virgin, Enhancing the teaching of structural dynamics using additive manu-
[19] S. Shun, N.A. Ahmed, Rapid prototyping of aerodynamics research models, Appl. facturing, Eng. Struct. 152 (2017) 750–757.
Mech. Mater. 217–219 (2012) 2016–2025. [49] M.V. OL, Water tunnel velocimetry results for the 1303 UCAV configuration, 24th
[20] K. Strelec J, C. Lagoudas D, Fabrication and testing of a shape memory alloy ac- AIAA Applied Aerodynamics Conference, AIAA, 2006AIAA 2006-2990.
tuated reconfigurable wing, SPIE's 9th Annual International Symposium on Smart [50] N.A. Ahmed, J.R. Page, Manufacture of an unmanned aerial vehicle (UAV) for
Structures and Materials, SPIE, San Diego, California, United States, 2002SPIE advanced project design using 3D printing technology, in: H. Zhao, K. Liu, X. Yu
4701. (Eds.), Advanced Design and Manufacturing Technology Iii, Pts 1-4, Trans Tech
[21] Y.A. Azarov, V.D. Vermel, A.V. Kornushenko, M.M. Novikov, S.V. Kamaev, Publications Ltd, Stafa-Zurich, 2013, pp. 970–980.
M.A. Markov, Experience in laser stereolithography and its application in manu- [51] T.V. Marcos, Martos JoF, R.G. Vilela, PGdP. Toro, A. Oliveira, I.D. Rego, 3D
facturing wind-tunnel aerodynamic models of various purposes, Seventh prototyping model design for experimental investigation in hypersonic shock
International Conference on Laser and Laser-Information Technologies, SPIE, tunnel, 21st AIAA International Space Planes and Hypersonics Technologies
Suzdal, Russia, 2002, pp. 433–440 SPIE 4644. Conference, AIAA, 2017AIAA 2017-2123.
[22] A.L. Heyes, D.A.R. Smith, Rapid technique for wind-tunnel model manufacture, J. [52] Id Rêgo, T.V. Marcos, V.A. Galvão, R.G. Vilela, M.A. Minucci, J.F. Martos, et al.,
Aircr. 41 (2004) 413–415. Preliminary studies on hypersonic flows over 3D printed models, 22nd AIAA
[23] I. Gibson, D. Rosen, B. Stucker, Additive Manufacturing Technologies: 3D Printing, International Space Planes and Hypersonics Systems and Technologies
Rapid Prototyping, and Direct Digital Manufacturing, Springer, 2015. Conference, AIAA, Orlando, FL, 2018AIAA 2018-5321.
[24] T. Wohlers, T. Caffrey, Wohlers Report 2012, (2012). [53] Z. Zhou, D. Li, J. Zeng, Z. Zhang, Rapid fabrication of metal-coated composite

13
W. Zhu Progress in Aerospace Sciences 110 (2019) 100541

stereolithography parts, Proc. IME B J. Eng. Manufact. 221 (2007) 1431–1440. [80] M. Fujino, H. Oyama, H. Omotani, Flutter characteristics of an over-the-wing
[54] Z. Zhou, D. Li, Z. Zhang, J. Zeng, Design and fabrication of a hybrid surface- engine mount business-jet configuration, 44th AIAA/ASME/ASCE/AHS/ASC
pressure airfoil model based on rapid prototyping, Rapid Prototyp. J. 14 (2008) Structures, Structural Dynamics, and Materials Conference, AIAA, Norfolk,
57–66. Virginia, 2003AIAA 2003-1942.
[55] W.J. Zhu, D.C. Li, Z.Y. Zhang, K. Ren, X.L. Zhao, D.G. Yang, et al., Design and [81] L.W. Traub, Effect of rapid-prototyped airfoil finish on loading at low Reynolds
fabrication of stereolithography-based aeroelastic wing models, Rapid Prototyp. J. numbers, J. Aircr. 50 (2013) 307–311.
17 (2011) 298–307. [82] J. Bartley-Cho, J. Henderson, Design and analysis of HiLDA/AEI aeroelastic wind
[56] D. Yang, Z. Zhang, Y. Sun, W. Zhu, A preliminary design and manufacturing study tunnel model, 26th AIAA Applied Aerodynamics Conference, AIAA, Honolulu,
of hybrid lightweight high-speed wind-tunnel models, Rapid Prototyp. J. 17 Hawaii, 2008AIAA 2008-7191.
(2011) 45–54. [83] B.K.S. Woods, L. Parsons, A.B. Coles, J.H.S. Fincham, M.I. Friswell, Morphing
[57] W. Zhu, X. Zhao, W. Zhang, K. Ren, Z. Zhang, D. Li, Design and evaluation of fully elastically lofted transition for active camber control surfaces, Aero. Sci. Technol.
configured models built by additive manufacturing, AIAA J. 52 (2014) 55 (2016) 439–448.
1441–1451. [84] Q. Ai, P.M. Weaver, M. Azarpeyvand, Design and mechanical testing of a variable
[58] W. Zhu, X. Zhang, D. Li, Flexible all-plastic aircraft models built by additive stiffness morphing trailing edge flap, J. Intell. Mater. Syst. Struct. 29 (2018)
manufacturing for transonic wind tunnel tests, Aero. Sci. Technol. 84 (2019) 669–683.
237–244. [85] S. Junk, W. Schröder, S. Schrock, Design of additively manufactured wind tunnel
[59] S. Daneshmand, R. Adelnia, S. Aghanajafi, Design and production of wind tunnel models for use with UAVs, Procedia CIRP 60 (2017) 241–246.
testing models with selective laser sintering technology using glass-reinforced [86] R. Jamieson, New options in aerodynamic modelling: CAD, CFD, stereo-
nylon, Mater. Sci. Forum 532–533 (2006) 653–656. lithography and shading in a workstation environment, Computer-aided Eng. J. 7
[60] S. Daneshmand, A.A. Nadooshan, C. Aghanajafi, Investigation of layer thickness (1990) 75–79.
and surface roughness on aerodynamic coefficients of wind tunnel RP models, in: [87] W. Zhu, D. Li, Z. Zhang, K. Ren, X. Zhao, D. Yang, et al., Design and fabrication of
C. Ardil (Ed.), Conference of the World Academy of Sciencee, Engineering and stereolithography-based aeroelastic wing models, Rapid Prototyp. J. 17 (2011)
Technology: World Acad Sci, Eng & Tech-Waset, 2007, pp. 7–12 WASET 2007-26. 298–307.
[61] A.A. Nadooshan, S. Daneshmand, C. Aghanajafi, Application of RP technology [88] M. Kampchen, A. Dafnis, H.G. Reimerdes, G. Britten, J. Ballmann, Dynamic aero-
with polycarbonate material for wind tunnel model fabrication, in: C. Ardil (Ed.), structural response of an elastic wing model, J. Fluids Struct. 18 (2003) 63–77.
Conference of the World Academy of Sciencee, Engineering and Technology, [89] C. Wang, G. Yin, Z. Zhang, S. Wang, T. Zhao, Y. Sun, et al., Design and fabrication
World Acad Sci, Eng & Tech-Waset, Bangkok, THAILAND, 2007, pp. 1–6 WASET of an aircraft static aeroelastic model based on rapid prototyping, Rapid Prototyp.
2007-26. J. 21 (2015) 34–42.
[62] S. Daneshmand, C. Aghanajafi, Description and modeling of the additive manu- [90] H. Korsch, A. Dafnis, H.G. Reimerdes, Dynamic qualification of the HIRENASD
facturing technology for aerodynamic coefficients measurement, Strojniski Vestn- elastic wing model, Aero. Sci. Technol. 13 (2009) 130–138.
J. Mech. Eng. 58 (2012) 125–133. [91] AgustaWestland, 1:8 Scale Wind Tunnel Model of the European Tiltrotor, CRP
[63] S. Daneshmand, C. Aghanajafi, H. Shahverdi, Investigation of rapid manufacturing Group, 2008, https://www.crp-group.com/PDF/cs-scale-wind-tunnel-model-
technology with ABS material for wind tunnel models fabrication, J. Polym. Eng. tiltrotor.pdf.
32 (2012) 575–584. [92] P.J. Penna, J.C. Su, G.F. Syms, Experimental and numerical studies of the effects of
[64] S. Daneshmand, C. Aghanajafi, H. Shahverdi, Investigation of papid manu- upper surface roughness on aileron performance, FAA In-Flight Icing/Ground De-
facturing technology effect on aerodynamics properties, Teh. Vjesn. 20 (2013) icing International Conference & Exhibition, SAE International, Chicago, Illinois,
425–433. 20032003-01-2101.
[65] A. De Gaspari, S. Ricci, L. Riccobene, Design, manufacturing and wind tunnel test [93] C. Aghanajafi, S. Daneshmand, A.A. Nadooshan, Influence of layer thickness on
of a morphing wing based on compliant structures, 24th AIAA/AHS Adaptive the design of rapid-prototyped models, J. Aircr. 46 (2009) 981–987.
Structures Conference, AIAA, San Diego, California, USA, 2016AIAA 2016-1316. [94] A.A. Nadooshan, S. Daneshmand, C. Aghanajafi, Evaluation of surface finish affect
[66] A. De Gaspari, L. Riccobene, S. Ricci, Design, manufacturing and wind tunnel on aerodynamic coefficients of wind tunnel RP models production with ABSi
validation of a morphing compliant wing, J. Aircr. 55 (2018) 2313–2326. material, Mater. Sci. Forum 594 (2008) 255–260.
[67] Z.W. Teo, T.H. New, T. Pfeiffer, S. Li, Z.A. Ong, B. Nagel, et al., Wind tunnel testing [95] E.T. Schairer, R.D. Mehta, M.E. Olsen, Effects of pressure-sensitive paint on ex-
of a joined-wing aircraft model with additive manufactured components, in: perimentally measured wing forces and pressures, AIAA J. 40 (2002) 1830–1838.
C.K. Chua, W.Y. Yeong, M.J. Tan, E. Liu, S.B. Tor (Eds.), 2nd International [96] S. Junk, W. Schröder, Application of Sustainable Design in Additive Manufacturing
Conference on Progress in Additive Manufacturing, Research Publishing, of an Unmanned Aerial Vehicle. Sustainable Design and Manufacturing 2016,
Singapore, 2016, pp. 170–175 10220-41827. Springer International Publishing, Cham, 2016, pp. 375–385.
[68] D.B. Landrum, Robert M. Beard, Paul A. LaSarge, von Sprecken, Natalie, [97] R. Udroiu, F. Dogaru, Rapid manufacturing of parts for wind tunnel testing using
Evaluation of stereolithography rapid prototyping for low speed airfoil design, polyjet technology, in: B. Katalinic (Ed.), Annals of DAAAM for 2009 &
35th Aerospace Sciences Meeting and Exhibit, AIAA, Reno, NV, 1997AIAA 1997- Proceedings of the 20th International Daaam Symposium, DAAAM International
719. Vienna, Wien, 2009, pp. 581–582 ISI BRC23.
[69] C.S. Cornelius, D.A. Gibson, Low-cost supersonic missile inlet fabrication tech- [98] C. Aghanajafi, S. Daneshmand, Integration of three-dimensional printing tech-
nique, Second International Conference on Intelligent Processing and nology for wind-tunnel model fabrication, J. Aircr. 47 (2008) 2130–2135.
Manufacturing of Materials, vol. 321, IEEE, Honolulu, HI, USA, 1999, pp. 325–329 [99] A.A. Nadooshan, S. Daneshmand, C. Aghanajafi, Application of RP technology
INSPEC 6409414. with polycarbonate material for wind tunnel model fabrication, Int. J. Mech.
[70] A.M. Springer, Application of Rapid Prototyping Methods to High-Speed Wind Aerosp. Ind. Mechatron. Manuf. Eng. 8 (2007) 371–376.
Tunnel Testing, NASA, 1998 NASA/TP-1998-208396. [100] S. Daneshmand, A.A. Nadooshan, C. Aghanajafi, Evaluation of FDM rapid proto-
[71] R.J. Hildebrand, R.C. Eidson, C. Tyler, Development of a low cost, rapid prototype, typing with ABSi material for airfoil design, in: S.J. Hwang, S.Y. Lee (Eds.),
lambda wing-body wind-tunnel model, 21st Applied Aerodynamics Conference. Advanced Manufacture: Focusing on New and Emerging Technologies, Trans Tech
Orlando, Florida, 2003 AIAA 2003-3818. Publications Ltd, Durnten-Zurich, 2008, pp. 249–254.
[72] C. Tyler, W. Braisted, J. Higgins, Evaluation of rapid prototyping technologies for [101] T. Bykerk, Sydney Researcher 3D Prints for Hypersonic Aircraft Testing, Beijing
use in wind tunnel model fabrication, 43rd AIAA Aerospace Sciences Meeting and Tiertime Technology Co. Ltd., 2018, https://www.tiertime.com/sydney-
Exhibit, AIAA, Reno, NV, United states, 2005, pp. 5081–5088. AIAA 2005-1301. researcher-3d-prints-for-hypersonic-aircraft-testing/.
[73] C. Tyler, Richard J. Schwartz, G. Fleming, S. Fonov, Rapid Prototyping-Unmanned [102] Z.W. Teo, T.H. New, S.Y. Li, T. Pfeiffer, B. Nagel, V. Gollnick, Wind tunnel testing
Combat Air Vehicle (UCAV)/sensorcraft. Force USA, (2008) AFRL-RB-WP-TR- of additive manufactured aircraft components, Rapid Prototyp. J. 24 (2018)
2008-3034. 886–893.
[74] C. Tyler, A joint computational fluid dynamics and experimental fluid dynamics [103] J.D. Hammond, Rapid wind tunnel model manufacture and model change tech-
test program, 20th International Congress on Instrumentation in Aerospace niques, 22th International Congress of the Aeronautical Sciences, Optimage Ltd.,
Simulation Facilities, IEEE, Reno, NV, United states, 2004, pp. 11270–11279 IEEE Harrogate, UK, 2000ICAS 2000-209.
8024308. [104] C.S. Cornelius, A Novel Application of Rapid Prototyping Techniques to Reduce
[75] D.A. Gebbie, M.F. Reeder, C. Tyler, V. Fonov, J. Crafton, PSP-based experimental Production Cost of Geometrically Complex Supersonic Missile Inlets, The
investigation of a blended wing body aircraft. 23rd AIAA Applied Aerodynamics University of Alabama in Huntsville, 1999 UMI 1396908.
Conference, AIAA, 2005AIAA 2005-4719. [105] I.S. Rego, T.V.C. Marcos, D.R. Pinto, R.G.S. Vilela, V.A.B. Galvao, A. Pivetta, et al.,
[76] A.M. Pankonien, G.W. Reich, N. Lindsley, B. Smyers, 3D-Printed wind tunnel Ground experimentation with 3D printed scramjet inlet models at hypervelocities,
flutter model, 58th AIAA/ASCE/AHS/ASC Structures, Structural Dynamics, and Aero. Sci. Technol. 55 (2016) 307–313.
Materials Conference, AIAA, Grapevine, Texas, 2017AIAA 2017-0411. [106] D. Alderfer, P. Danehy, J. Inman, K. Berger, G. Buck, R. Schwartz, Fluorescence
[77] A.M. Pankonien, R. Durscher, N.D. Bhagat, G. Reich, Multi-material printed visualization of hypersonic flow over rapid prototype wind-tunnel models, 45th
trailing edge control surface for an aeroservoelastic wind tunnel model, 2018 AIAA Aerospace Sciences Meeting and Exhibit, AIAA, 2007AIAA 2007-1063.
Multidisciplinary Analysis and Optimization Conference, AIAA, 2018AIAA 2018- [107] P.M. Danehy, D.W. Alderfer, J.A. Inman, K.T. Berger, G.M. Buck, R.J. Schwartz,
2930. Fluorescence imaging and streakline visualization of hypersonic flow over rapid
[78] M.F. Reeder, W. Allen, J.M. Phillips, R. Dimmick, Wind-tunnel measurements of prototype wind-tunnel models, Proc. Inst. Mech. Eng. Part G-J. Aerosp. Eng. 222
the E-8C modeled with and without winglets, J. Aircr. 45 (2007) 345–348. (2008) 637–651.
[79] C. Black, K.V. Singh, S. Goodman, A. Altman, R. Kolonay, Design, fabrication and [108] R.B. Torres, G.B. Jacobs, M.J. Cave, Experimental study on the use of synthetic jet
testing of 3D printed wings for rapid evaluation of aeroelastic performance, 2018 actuators for lift control, SME Turbo Expo 2016: Turbomachinery Technical
AIAA/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference, Conference and Exposition, ASME, Seoul, South Korea, 2016, pp. GT2016–56403.
AIAA, Kissimmee, Florida, 2018, pp. 1–17. AIAA 2018-1997. [109] K. Fujii, Progress and future prospects of CFD in aerospace - wind tunnel and

14
W. Zhu Progress in Aerospace Sciences 110 (2019) 100541

beyond, Prog. Aero. Sci. 41 (2005) 455–470. Making Same, The Johns Hopkins University, United States, 2004 10/130813.
[110] W. Zhu, K. Miao, D. Li, Aeroelastic models with integrated stiffness-contributing [114] G.D. Goh, S. Agarwala, G.L. Goh, V. Dikshit, S.L. Sing, W.Y. Yeong, Additive
shell structures built by additive manufacturing, Eng. Struct. 187 (2019) 352–361. manufacturing in unmanned aerial vehicles (UAVs): challenges and potential,
[111] M. French, F.E. Eastep, Aeroelastic model design using parameter identification, J. Aero. Sci. Technol. 63 (2017) 140–151.
Aircr. 33 (1996) 198–202. [115] P. Marks, 3D printing takes off with the world's first printed plane, New Sci. 211
[112] S. Raja, S. Rajappa, S. Janardhanam, Aeroelastic Testing of Aerospace Vehicles - (2011) 17–18.
Experiences Gained during Four Decades National Conference on Wind Tunnel [116] M.M. Stern, E. Cohen, VAST AUAV (variable airspeed telescoping additive un-
Testing (2009-IT11). IISc, Bangalore, India, National Aerospace Laboratories, manned air vehicle), RAPID 2013 Conference and Exposition, Including 3D
India, 2009 NAL 5170. Imaging, SME, Pittsburgh, PA, 2013RAPID 2013.
[113] R.R. Heisler, C.L. Ratliff, Rapid Prototype Wind Tunnel Model and Method of

15