Aerospace Science and Technology 65 (2017) 54–61

Contents lists available at ScienceDirect

Aerospace Science and Technology

www.elsevier.com/locate/aescte

Optimization of solar-powered hybrid airship conceptual design

Lanchuan Zhang, Mingyun Lv ∗ , Junhui Meng, Huafei Du
School of Aeronautic Science and Engineering, Beihang University, Beijing, China

a r t i c l e i n f o a b s t r a c t

Article history: This paper aims to develop a high-altitude solar-powered hybrid airship, which combines aerodynamic
Received 12 November 2016 lift and buoyancy with buoyancy force and cruises continuously. A multi-lobed configuration is employed
Accepted 17 February 2017 and the solar irradiance and photovoltaic array model are proposed. After introducing all the parts mass
Available online 27 February 2017
of the hybrid airship and design procedure, an optimal problem with the constraint of energy and the
Keywords:
equilibrium between buoyancy and weight are proposed. Through a hybrid optimization search method,
Hybrid airship the solution is obtained and the sensitivities of geographical location and seasons are discussed. The
Solar array results show a comprehensive influence of the constraint both in latitude and wind field environment.
Conceptual design © 2017 Published by Elsevier Masson SAS.
Optimization

1. Introduction
Development of hybrid airships has been an important stream
in both the military and civil aviation industries for decades and
will continue to draw great attention in the future. The novel
projects develop rapidly in decades such as Lockheed Martin Sky-
Tug, Hybrid Air Vehicles Airlander 50 [1] which can carry dozens
of tons cargo for remote transportation. In addition, hybrid airship
as a high altitude platform can cruise and loiter in the stratosphere
for remote reconnaissance and ground surveillance mission [2,3].
Generally, the hybrid airship is a new configuration combining
static buoyancy with aerodynamic lift produced by lifting body and
vector thrust. One of the important differences from conventional
airship is an improved aerodynamic configuration which generates
a considerable amount of lift such as delta lifting body, winged
and multi-lobed shapes [4–9]. With a multi-lift source and high
lift–drag ratio, hybrid airship has an adjustable buoyancy ratio (BR)
which plays an important role distinct from the conventional one.
Based on an great improvement on changeable lift, hybrid air-
ship can adjust the altitude flexibly to overcome different weather,
wind field and mission requirement. And the performance is also
distinctly different among various configurations such as winged
and multi-lobed ones.
In addition, as a result of the development of regenerative
power lately, clean energies are gradually applied as more durable
and stable power for high altitude hybrid airship such as solar
energy which seems to be inexhaustible [10,11]. Regardless of air-
craft types, HAA [12] and HALE [13,14] can both achieve several
* Corresponding author at: School of Aeronautic Science and Engineering, Beihang
University, No. 37 Xueyuan Road, Haidian District, Beijing, China.
E-mail address: lv503@buaa.edu.cn (M. Lv).
days flight using photovoltaic array panels for propulsion power.
However, the efficiency of the flexible solar cell is only 8% [15],
though the rigid one has a nearly 30% efficiency [16]. And the
energy supply during the night is also a tough challenge for stor-
age [17]. It is noteworthy that the fuel cells have more than
1000 w h/kg [18] specific energy density than lithium battery with
a 200∼300 w h/kg, which will be a great potential application and
weight reduction in energy storage system during night [17]. With
an aerodynamic advantage and potential energy system, high alti-
tude hybrid airship design and manufacture are undergoing rapid
development accordingly.
A methodology for sizing of high altitude unconventional air-
ship was proposed by Ceruti [19], the methodology presented a
parametric approach considering the evolution in materials and
technology, new configurations, and modern power and demon-
strated a graphical and straightforward mathematical operations
to obtain the solutions. A Multi Disciplinary Optimization (MDO)
[20] was also presented with heuristic algorithms to optimize the
aerodynamic shape. A comparison among these algorithms has in-
dicated that the genetic algorithm (GA) and Particle Swarm Op-
timization(PSO) had better results than Simulated Annealing (SA)
and Monte-Carlo method. Song et al. [2] demonstrated a high al-
titude winged airship with solar array atop the body and wing.
A Non-dominated Sorted Genetic Algorithm optimization has been
carried out to determine the configuration parameters with differ-
ent cruise modes for day and night. Jenkins et al. [21] proposed
an optimization design method using a genetic algorithm (GA)
for High Altitude Reconnaissance Vehicle (HARVe) which can be
launched from a missile. The power generation system is optimized
subject to constraints on vehicle buoyancy and energy balance.
Alam et al. [22] presented a methodology for conceptual design
of a high altitude airship with an improved optimization process.

http://dx.doi.org/10.1016/j.ast.2017.02.016
1270-9638/© 2017 Published by Elsevier Masson SAS.
 L. Zhang et al. / Aerospace Science and Technology 65 (2017) 54–61
By coupling with other optimizations, the methodology sizes the
airship to obtain an optimal solution for some user-defined ob-
jective function. However, it was rare in these studies to discuss
optimization design for a high altitude solar-powered lifting-body
hybrid airship for different flight strategies.
This paper presents details of an optimization design and anal-
yses for initial sizing of solar powered hybrid airship with different
flight conditions. The rest of this paper include: a solar irradiance
and the geometric photovoltaic array on multi-lobed configuration
model is proposed. After introducing the mass distribution and the
design process, an optimization method was used to search the op-
timum solution of the minimum total mass as an objective. Then
the sensitivity analysis is established to research the effects of date
and latitude on the optimal mass of hybrid airship. The obtained
conclusions provide a valuable reference for the preliminary design
of high altitude hybrid airship.
2. Conceptual design
2.1. Solar irradiance model
Ignoring the sky scattering and earth radiation, the solar irradi-
ance I 0 at the stratospheric altitude can be calculated with:
 2
1 + e 0 cos λe
I = I0
1 − e 20
where I 0 is the solar constant that has a value of 1367 W/m2 , e 0 is
the orbital eccentricity, for earth e 0 = 0.0016708, the true anomaly
λe is given by:
λe = θday + 0.0344 sin(θday ) + 3.49 × 10−4 sin(2θday )
θday is the day angle of the sun and is given by:
θday = 2π ( N − n)/365.24
where N is the day number in a year which means N = 1 repre-
sents January 1st and N = 365 December 31st in a non-leap year.
And the sun radiation power density received by hybrid airship
is:
P sun = I sin β
where solar altitude β , the angle between the vector pointing to
the sun and its projection on horizontal plane is calculated by:
sin β = sin θ sin ϕ + cos θ cos ϕ cos ω(t )
where θ is latitude, ϕ is declination angle:
  
284 + n
ϕ = 23.45π sin 2π × /180
365
And the ω(t ) is solar hour angle, t is the time of day.
ω(t ) = π − π t /12
3. Photovoltaic array model
As for conventional shape a double ellipsoid and a cylinder in
the middle body to place the solar array, the multi-lobe shape can
be considered as several conventional bodies paralleled with the
photovoltaic array on the top surface as shown in Fig. 1.
Similar to conventional layout, the three portion of solar array
are installed on the upper surface of lobes in symmetric form. As-
suming the lobe is an ellipsoid shape, the curved envelope and
solar panel can be described as a rotating surface. The equation is
given by:
55
Fig. 1. Envelope and solar array configuration.
y 2 + z2 = r 2 (x) 0≤x≤L
Considering the sufficiently small grids as tilted planes instead
of the curved surface of solar panels, the length of each section is
dx along the direction of the rotation axis. And the width along
the span can be represented as the length of small circular arcs
along the circumferential direction. The area of tilted element i j is
defined as A i j , which can be described as [23,24]:

A i j = rdθ dx 1 + r  (x)2
where θ is the rotation angle along the circumferential direction.
To simplify the calculation model, assuming θ0 is the central
angle of solar array for each lobe and the solar array is only laying
on the top surface, l0 is the length of the solar array atop the hull,
the total area of solar array can be expressed by:

A array = N L Ai j
yi θj
where the N L is the number of lobes.
In the inertial coordinate system, the unit vector of solar di-
 s is a function of elevation angle and azimuth angle
rect radiation n
which can be expressed as:
 s = (cos θele cos θazi , − cos θele sin θazi , − sin θele )
n
While in the body reference frame, the incident solar radiation is
changed with the different position on the panel attached to the
envelope when the time is certain. Therefore, the solar direct radi-
ation on the elements i j can be expressed by
qi j = αproj · I · A i j
where αproj is the projection coefficient of solar direct radiation
on the grid normal vector through a coordinate transformation.
Assuming the azimuthal angle of the hybrid airship is zero, the
average solar irradiance per unit area is calculated using MATLAB
as shown in Fig. 2.
Considering the added mass of collector grid network and other
auxiliary components, the mass of the solar array can be expressed
as:
1
mpanel = ρpanel S panel = ρpanel α S ref R
2
where R is the ratio of the solar cell area atop the HA, α is a factor
which is only relevant to the shape of hybrid airship.
3.1. Fuel cell model
Energy management monitors and controls the charge and dis-
charge loops of the fuel-cells modules. Power conversion generates
56 L. Zhang et al. / Aerospace Science and Technology 65 (2017) 54–61

Fig. 2. Solar irradiance in different latitude and date.

Fig. 4. Power balance between day and night.

Fig. 3. Hybrid energy management.

area of hybrid airship, the motor and propeller mass can be ex-
pressed by:
the required voltage for the motor and avionics systems. Details of
the components of the energy system are presented in Fig. 3. P max
mmotor+propeller =
Fuel cells as a storage system combine fuel cell stacks, hydro- S P motor+propeller
gen storage container and attachments. The energy balance is per-
formed between the energy produced by the solar array and the ρair ( v max )3 C D S ref
=
energy consumed by the hybrid airship during the 24-h period as 2S P motor+propeller ηmotor+propeller
shown in Fig. 4, where Q1 represents the power available for stor-
age which is the sum power of Q2 and Q4 for night consumption. 3.3. Structure
Q3 is the power supplied for the power consuming function during
the daytime. At the summer solstice, the average required power The mass of the envelope and structure is computed from the
in a day for solar panels is approximatively 460 W per square me- wetted area and the specific mass:
ter.
The fuel cells are weighted by the nighttime endurance. As- menvelope = ρenvelope S wetted
suming the flight condition is straight and level with a constant where ρenvelope is the density of envelope with some modification
velocity, the mass of fuel cells is: in view of ballonets and internal structural frames. The mass of tail
E fuel-cells P required T night can be computed in a similar way using the following expressions:
mfuel-cells = =
ρfuel-cells ρfuel-cells mtail = ρtail S tail = λtail ρtail S ref
 
1 ρair v 2 S ref C D
= + P others T night The ratio of tail λtail is referred to the Lockheed Martin design
ρfuel-cells ηfuel-cells 2ηmotor+propeller [16]. Therefore, the total mass of the HA:
The ηfuel-cells represents the efficiency of fuel cells round-trip.
mtotal = mpanel + mbattery + mmotor+propeller + menvelope
3.2. Propulsion system + mtail + m payload .

In this design, propulsion system consists of motor propeller 3.4. Process of design
and accessories, the power requirement is in turn based on the
thrust power required from the propeller and the efficiency factor. As for a traditional design flow, weight balance is an initial rule
According to the maximum speed drag coefficient and reference for routine flight. In a steady level flight of hybrid airship, the total
 L. Zhang et al. / Aerospace Science and Technology 65 (2017) 54–61 57

Fig. 5. Design flow chart of hybrid airship.

weight is balanced by static buoyancy and aerodynamic lift with- The aerodynamic configuration design of a hybrid airship is
out vector thrust: more complicated than a traditional one which has been analyzed
in early study [25]. As for a solar-powered aircraft, the power ob-
1
mtotal g = B + L = (ρair − ρhe ) gVol + ρair v 2 S ref C L tained from upper surface solar cell panels is directly proportional
2 to the pave factor R and surface area, which, in turn, affects most
Aerodynamic lift and aerostatic force can be expressed as a of the aircraft performance parameters, such as lift, drag, weight,
fraction of total weight: and cruise speed. In addition, because of most of the design pa-
rameters interrelated, one parameter is adjusted usually causing
B = BR · mtotal g L = (1 − BR) · mtotal g the reevaluation or recomputation of all other parameters. Tradi-
tionally, iteration of certain design procedures is used to obtain a
Assuming that BR changes little during a continuous flight at a
compromise solution. The design procedures are usually tedious.
fixed level altitude.
The design flow chart is shown in Fig. 5.
The reference platform area S ref can be expressed as a multi-lobed
This process begins with an initial volume, cruise speed and ini-
shape:
tial BR, then the both static and aerodynamic lift and weight of the
propulsion and structure are computed, and the area of the solar
S ref = N L Vol2/3
cell panel and the mass of solar array and fuel cells are obtained.
Where Vol is the volume of the envelope. Therefore the total required power for hybrid airship operation and
58 L. Zhang et al. / Aerospace Science and Technology 65 (2017) 54–61

Fig. 6. Configuration of hybrid airship.

available power from panels are evaluated. If the supplied power is

lower than the consumed power, the solar panel area is increased.
If the sum of lift and buoyancy is less than the weight, the cruise
attack angle and initial volume is decreased but if the combination
is higher, the attack angle and volume is increased correspond-
ingly and the design process repeated. If the lift is equal to the
total weight of hybrid airship, the iteration is complete.
Fig. 6 shows the preliminary configuration view of the solar-
powered hybrid airship. The photovoltaic array is paved on the
upper surface of each lobe. And the distributed X-tails are installed
to ensure the stability and flight control not merely the inner bal-
lonets.

4. Optimization formulation

Fig. 7. Optimization process of hybrid method.

The aim of this section is to optimize the weight of multi-lobed
hybrid airship with a given payload so that the length and the
size of hull are as small as possible. After the parameterization,
the optimization process can be initiated to search for the opti-
mum solution. The object is to search for the optimal cruise speed,
referenced area and BR that satisfy the constraints and minimum
weight. The reference area is a dimension parameter sizing the air-
craft and the BR is ambivalent parameter between the lift and
buoyancy, or an airplane and airship. The cruise velocity deter-
mines the platform flight and wind resistance performance, further
directly influence the propulsion power and general parameters.
Therefore it has been set as the third variable that is restricted by
the local wind speed. Note that the stability and control of hybrid
airship are not taken into consideration during the optimization
process.
The optimization model is shown as follows.
Objective:
Min mtotal
Constraint (s.t):
g1:
B + L = (mpanel + mbattery + mmotor+propeller + menvelope
+ mtail + mpayload )g
g2:
Q sun P payload T day − P proplusion T day − P fuel−cells T night ≥ 0
g3:
0≤C L ≤C L max
Initial variables:
v ∈ ( v local , v max ) (m/s)
 
S ref ∈ (3000, 15000) m2
BR ∈ (0, 1)
where v local is the wind velocity for different local wind situation.
Optimization problems are solved using a hybrid algorithm con-
sisting of the genetic algorithm (GA) treated as global search algo-
rithm and the gradient-guided search method [26] used for opti-
mizing possible solution.
GA has been applied broadly in the aeronautic field [11,21,
27–30]. GA attempt to minimize the objective function by simulat-
ing the process of evolution. In the GA, the design points are called
genes, the objective is the fitness function. Unlike other optimiza-
tion techniques, GA is a population-based search algorithm, which
works with a population of strings or real variables that represent
different potential solutions. Therefore, GA has implicit parallelism
that enhances its search capability and the optimal solution can be
located more quickly for many complex problems.
Although GA is very efficient at exploring the entire search
space, its ability to find the precise location of the optimum solu-
tion was poor [31]. And other problems such as premature conver-
gence and uncertainty have also been serious in some situations.
Therefore, combining with other algorithms is a feasible way for
GA to solve these problems [26,32–34]. Considering the gradient
method is a traditional optimization algorithm with quick conver-
gence and high accuracy, the gradient-guided search method has
a high efficiency to search the neighbor extremum. Therefore, we
propose a hybrid optimizer that takes advantage of the GA’s abil-
ity to perform a global search and gradient method to efficiently
find the nearest local optimum and enforce constraints. The block
diagram of the hybrid optimizer is shown in Fig. 7.
Finally, The following results are obtained after the completion
of the optimization search routine. The minimum mass is 19946 kg
with v = 19.7 m/s, S ref = 8829 m2 and BR = 0.615 at 40◦ north
latitude on the summer solstice. The optimal results are listed in
Table 1 and the weight components are shown in Fig. 8. It can be
seen that the mass of structure subsystem and energy subsystem
are the most part of the total mass.
 L. Zhang et al. / Aerospace Science and Technology 65 (2017) 54–61 59

Table 1
Optimal results of hybrid airship.

Variable Value
v 19.7 m/s
S ref 8829
BR 0.615
mpaylaod 1000 kg
mfuelcell 10396 kg
msolar array 2589 kg
mproplusion 1600 kg
mstructure 4061 kg
mtail 300 kg

5. Results and discussion

Fig. 8. The mass distribution of the optimization result.

Assuming the flight altitude of airship is 20 km during the

simulation process. Due to the local wind speed as an initial con-
straint in the iteration process, the wind field profile becomes an
essential requirement for optimization design. Based on the NASA
wind field investigation [35], the wind magnitude and direction
at different elevation at five coastal observation stations (Port-
land, Maine 43.4◦ , Washington DC 38.53◦ , Charleston SC 32◦ , Cape
Kennedy Florida 28.5◦ , Miami Florida 25.46◦ ) on the East Coast of
the United States were obtained. And the mean wind and the 95
percentile of wind value by interpolation are shown in Fig. 9.
An important influence of seasons and latitude on the results
of optimization is displayed in Figs. 10–11. Note that all other pa-
rameters are kept the same.
As is shown in Figs. 10–11, the designed cruise velocity and BR
have the same tendency with the optimal total weight of hybrid
airship for different seasons. As results of varying local wind speed
and solar irradiance condition which have distinct effects on opti-
mization solutions, a wind variation trend is not indicated a higher
cruise velocity and total weight.

Fig. 9. Wind speed at 20 km in different latitude (a)average value (b) 95% level value.

Fig. 10. Results sensitivity at different latitude in spring and summer.

60 L. Zhang et al. / Aerospace Science and Technology 65 (2017) 54–61
Fig. 11. Results sensitivity at different latitude in autumn and winter.
As a result of the day going on from the summer solstice to
autumn equinox in Fig. 11(a), although the solar irradiance has
been decreased, and with a descending tendency of wind speed
as latitude escalating in autumn, the wind speed constraint is also
decreased from low latitude to high latitude, which leads to a sim-
ilar trend of design parameters and results, both in summer and
autumn.
As shown in Fig. 11(b), the optimal total weight is quite lower
than other seasons in winter. The result is probably due to the low
level total solar irradiation is one of the crucial reason. Although
solutions which did not match the constraints at higher latitude
are discarded, a increasing trend for the optimum variables and
total weight is obvious with latitude escalating. The BR, as a crucial
design parameter leads the important effects on the final result, it
can be demonstrated that a 0.6 ∼ 0.7 value is appropriate for this
hybrid airship design except in winter where a higher wind speed
and lower available solar energy result in a higher BR for more
static buoyancy.
6. Conclusion
A conceptual design process has been put out for a high-
altitude solar-powered hybrid airship, after demonstrated solar ir-
radiance model on photovoltaic arrays attached to the multi-lobed
shape, the components of mass expression and the design flow
of hybrid airship have been presented. An hybrid optimization of
genetic algorithm/gradient-based method was employed to search
the optimum solution in the design space. The results show that
the cruise velocity, envelope volume and BR with a 1000 kg pay-
load capacity at 20 km altitude are 19.7 m/s, 159656 m3 , 0.615
respectively.
And effects of geography and time sensitivity are discussed. The
results show that the optimal total weight varies with trend in
buoyancy ratio and cruise speed over different latitudes and dates.
With other parameter unaltered and a fixed payload 1000 kg, the
results of minimum weight in winter are inferior to other seasons.
The optimization model and results are of reference value for the
configuration design in low speed stratospheric hybrid airship.
Conflict of interest statement
None declared.
References
[1] B.E. Prentice, R. Knotts, Cargo airships: international competition, J. Transp.
Technol. 4 (2014) 187–195.
[2] H. Wang, B. Songand, X. Zhong, Configuration design and sizing optimization of
a winged airship, in: Configuration Design and Sizing Optimization of a Winged
Airship. International Conference on Network Computing and Information Se-
curity, 2011, pp. 41–45.
[3] B. Brandon, The suitability of hybrid vs. conventional airships for persistent
surveillance missions, in: 48th AIAA Aerospace Sciences Meeting Including the
New Horizons Forum and Aerospace Exposition, 2015/11/15, American Insti-
tute of Aeronautics and Astronautics, 2015.
[4] M.D. Ardema, Feasibility of modern airships: preliminary assessment, J. Aircr.
14 (11) (1977) 1140–1148.
[5] G. Carichner, L.M. Nicolai, Hybrids...the Airship Messiah?, in: AIAA Lighter-
Than-Air Systems Technology Conference, 2013.
[6] J. Agte, T. Gan, F. Kunzi, Conceptual, design of a hybrid lift airship for intra-
regional flexible access transport, in: 48th AIAA Aerospace Sciences Meeting
Including the New Horizons Forum and Aerospace Exposition, Orlando, Florida,
January 2010, 2010.
[7] R. Jensen, Operational considerations for large hybrid air vehicles, in: AIAA Avi-
ation Technology, Integration and Operations Forum, September 2004, 2004.
[8] A.D. Andan, W. Asrarand, A.A. Omar, Investigation of aerodynamic parameters
of a hybrid airship, J. Aircr. 49 (2) (2012) 658–662.
[9] A. Ghaffar, The Development of a Mathematical Model of a Hybrid Airship,
Dissertations & Theses – Gradworks, 2012.
[10] A. Colozza, J. Dolce, Initial Feasibility Assessment of a High Altitude Long En-
durance Airship, 2004.
[11] J.K. Shiau, D.M. Ma, C.W. Chiu, Optimal sizing and cruise speed determination
for a solar-powered airplane, J. Aircr. 47 (2) (2010) 622–629.
[12] G.D. Modica, T. Nehrkornand, T. Myers, An investigation of stratospheric winds
in support of the high altitude airship.
[13] M.J. Patil, D.H. Hodgesand, C.E.S. Cesnik, Nonlinear aeroelasticity and flight dy-
namics of high-altitude long-endurance aircraft, J. Aircr. 38 (1) (2001) 88–94.
[14] E. Cestino, Design of solar high altitude long endurance aircraft for multi pay-
load & operations, Aerosp. Sci. Technol. 10 (6) (2006) 541–550.
[15] Q. Chen, M. Zhuand, K. Sun, Analysis to effects on conceptual parameters of
stratospheric airship with specified factors, J. Comput. 6 (5) (2011) 1055–1062.
[16] G.E. Carichner, L.M. Nicolai, Fundamentals of Aircraft and Airship Design, vol. II,
Airship Design, American Institute of Aeronautics and Astronautics, AIAA, 2013.
[17] G.A. Khoury, Airship Technology, Cambridge University Press, 2012.
[18] G. Romeo, F. Borello, G. Correa ENFICA-FC, Design of transport aircraft powered
by fuel cell & flight test of zero emission 2-seater aircraft powered by fuel cells
fueled by hydrogen, Int. J. Hydrog. Energy 38 (1) (2013) 469–479.
[19] A. Ceruti, P. Marzocca, Conceptual approach to unconventional airship design
and synthesis, J. Aerosp. Eng. 27 (6) (2013) 04014035.
[20] A. Ceruti, V. Voloshinand, P. Marzocca, Multi-disciplinary design optimiza-
tion of unconventional airship configurations with heuristic algorithms, in:
AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Con-
ference, April 2013, 2013.
[21] J. Jenkins, J. Samsundarand, V. Neradka, A design methodology for optimal
power generation in high altitude airships using genetic algorithms, in: Inter-
national Energy Conversion Engineering Conference, August 2005, 2005.
[22] M.I. Alam, R.S. Pant, A methodology for conceptual design and optimization of
a high altitude airship, AIAA J. (2013).
[23] J. Li, M. Lvand, K. Sun, Optimum area of solar array for stratospheric solar-
powered airship, Proc. Inst. Mech. Eng., G J. Aerosp. Eng. (2016).
[24] J. Li, M. Lv, K. Sun, Thermal insulation performance of lightweight substrate for
solar array on stratospheric airships, Appl. Therm. Eng. 107 (2016) 1158–1165.
 L. Zhang et al. / Aerospace Science and Technology 65 (2017) 54–61
[25] P. Mi, J. Mengand, M. Lyu, Aerodynamic and overall parameters analysis
of buoyancy-lifting hybrid airship, Beijing Hangkong Hangtian Daxue Xue-
bao/Journal of Beijing University of Aeronautics & Astronautics 41 (6) (2015)
1108–1116.
[26] A. Bos, Aircraft conceptual design by genetic/gradient-guided optimization, Eng.
Appl. Artif. Intell. 11 (3) (1998) 377–382.
[27] Y. Dai, J. Wangand, G. Lin, Parametric optimization and comparative study of
organic Rankine cycle (ORC) for low grade waste heat recovery, Energy Convers.
Manag. 50 (3) (2009) 576–582.
[28] L.F. Gonzalez, Robust Evolutionary Methods for Multi-Objective and Multdisci-
plinary Design Optimisation in Aeronautics, University of Sydney, 2005.
[29] O. Chernukhin, D.W. Zingg, Multimodality and global optimization in aerody-
namic design, AIAA J. 51 (51) (2013) 1342–1354.
61
[30] I.A. Chaudhry, A. Ahmed, Preliminary aircraft design optimization using genetic
algorithms, Am. J. Sci. Ind. Res. 73 (5) (2015) 302–307.
[31] M.A. Abido, Optimal design of power-system stabilizers using particle swarm
optimization, IEEE Power Eng. Rev. 17 (3) (2002) 406–413.
[32] D. Bos, Multidisciplinary design optimization of a supersonic transport aircraft
using a hybrid genetic/gradient-based algorithm, AIAA J. (2013).
[33] A. Joodaki, A. Ashrafizadeh, Surface shape design in fluid flow problems via
hybrid optimization algorithms, Aerosp. Sci. Technol. 39 (2014) 639–651.
[34] L. Yang, Y.U. Suyuan, G. Yang, HTGR Helium Turbine Conceptual Design by Ge-
netic/Gradient Optimization, 2004.
[35] T. Strganac, Wind study for high altitude platform design, AIAA J. (2013).