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

net/publication/236660533

Design Optimization of a 3-Unit Satellite De-Orbiting Mechanism

Conference Paper · June 2012

CITATIONS READS

0 635

3 authors:

Pinar Acar Melike Nikbay

Virginia Polytechnic Institute and State University Istanbul Technical University
90 PUBLICATIONS   288 CITATIONS    91 PUBLICATIONS   644 CITATIONS   

SEE PROFILE SEE PROFILE

A.R. Aslan
Istanbul Technical University
80 PUBLICATIONS   175 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Optimization of curvilinear and spatially varying fiber paths in composite structures View project

Development of BEEAGLESAT and HAVELSAT View project

All content following this page was uploaded by Melike Nikbay on 22 January 2015.

The user has requested enhancement of the downloaded file.

 DESIGN OPTIMIZATION OF A 3-UNIT SATELLITE DE-ORBITING MECHANISM

Pınar Acar (1), Melike Nikbay (2), Alim Rüstem Aslan (3)
(1,2,3)
Istanbul Technical University, Faculty of Aeronautics and Astronautics, Astronautical
Engineering Department
acarpin@itu.edu.tr, nikbay@itu.edu.tr, aslanr@itu.edu.tr

ABSTRACT

Design optimization of a cost effective mechanical de-orbiting system based on aerodynamic drag
enhancement principle is presented. The de-orbiting mechanism consists of spiral springs as energy
producing mechanism for mechanical deployment as well as thin membranes which are used to
provide orbit decay as aero-brake structures. The optimization problem aims to maximize
aerodynamic drag force and moment produced by spiral spring while introducing design variables
which are dimensions of membrane structure and spiral spring. Calculation of aerodynamic drag
force even considers minimum values of orbital tangential velocity and effective satellite surface
area for reliability. The design is further constrained to restrict the membrane mass and forces acted
on the spiral spring so as to maintain structural integrity and safety. The optimum design is selected
through pareto optimal designs and its orbital life-time is computed in Satellite Tool Kit (STK). The
maximum de-orbiting time is determined as 25 years, which is the maximum orbital life time
suggestion of United Nations Inter-Agency Space Debris Coordination Committee (IADC),
however the main purpose of the present study is to provide a possible shortest time de-orbiting.

1 INTRODUCTION

One of the highly appreciable issues in Low Earth Orbit (LEO) space investigations of today’s
world is day-by-day growing space debris danger. Man-made objects with 10 cm or bigger
dimensions threaten the international space commerce seriously [1], thus restrictions in a global
sense are required to put into practice to prevent collisions. United Nations Inter-Agency Space
Debris Coordination Committee (IADC) suggests 25 years of maximum orbital lifetime in 2002 [2].
Satellite de-orbiting mechanisms have been attracting high attention in recent years since they
provide faster orbit decay. During orbital motion, all satellites lose their energies due to atmospheric
drag, however the orbit decay time is much more than IADC's suggestion. Orbital life-time can be
thousands of years above 500 km altitude [3] and must be reduced by making use of de-orbiting
devices. Satellite de-orbiting systems step up the orbit decay process by benefiting from various
physical principles. Propulsion system, electrodynamic tether drag augmentation, aerodynamic drag
enhancement, solar sails and plasma brake device are the prevalent approaches for de-orbiting in
LEO. De-orbiting systems with electrodynamic tether are useful options for CubeSats whose mass
and volume sources are so limited [3, 4]. In the work of Corsi [5], de-orbiting is based on
electrodynamic drag produced by Lorentz force which is generated from the interaction between the
current in wire and geomagnetic field. Lappas [6] presented 3-Unit (3U) CubeSail which uses sail
membrane as a drag producing mechanism. There are also standalone devices for de-orbiting a
satellite in LEO such that CubeSat Deorbit and Recovery System (DRS) was developed by
Andrews Space [7] and a plasma brake device based on Coulomb Drag Interaction between
ionospheric plasma and negatively charged tether with relatively less energy and mass requirements

The 4S Symposium 2012 – P. Acar, M. Nikbay and A. R. Aslan 1

was presented by Janhunen [8].

Aerodynamic drag enhancement principle, which is previously studied in the works of Maessen [9]
and Lokcu [10], makes use of surface area augmentation and is preferred in this work within
context of a deployable mechanical de-orbiting system. The surface area is directly proportional to
the aerodynamic drag force acted to the satellite. A higher value of aerodynamic drag indicates a
faster decrease in orbital altitude. The minimum border of this altitude, where the satellite burns out
and vanishes, is ascertained as 180 km in the present work. The de-orbiting mechanism consists of
membrane structures with assistive steel beams as drag producing mechanism as well as spiral
springs to generate the necessary deployment moment. The main concerns of this work are to
maximize both the aerodynamic drag force and spring moment for an easier deployment and faster
orbit decay. Therefore, multi-objective design optimization is performed by varying input
parameters which are thickness, width and outside diameter of the spiral spring, and length and
width of the membranes. Mass constraint is defined to prevent extreme augmentation of the
membrane area while the stress due to acted forces on the spiral spring forms the second constraint.
Finally, orbital life-time, which is basically affected from initial orbit, cross-sectional area, mass
and drag coefficient [10], is estimated through STK (Satellite ToolKit). The maximum de-orbiting
time is determined as 25 years in the present work by considering the maximum orbital life-time
suggestion of IADC. Optimum mechanism is finally checked to ensure whether it satisfies the
maximum life-time criterion.

2 DESIGN PROPERTIES OF 3USAT DE-ORBITING MECHANISM

The de-orbiting mechanism of interest in the present work is a derivation of a designed model in
Space Systems Design and Testing Laboratory (SSDTL) of Istanbul Technical University (ITU)
within the context of a nano-satellite development project, 3USAT, in collaboration with
TURKSAT Inc. of Turkey. 3USAT's de-orbiting device is located between camera and modem. All
sub-systems of the satellite are shown in Figure 1.

The fundamental elements of the de-orbiting system are spiral springs for deployment energy
production as well as thin membranes which behave like aero-brake structures for orbit decay. The
membranes and spiral springs are made of kapton and steel respectively. The material properties are
summarized in Table 1. Moreover, the de-orbiting mechanism contains a storing unit with storing
cylinders which are used for membrane storage when the system is de-active and bands for energy
transmission from springs to storage cylinders. Lock mechanism prevents the system from
undesirable deployment. The general representation of the considered de-orbiting mechanism with
all system components is given in Figure 2 and 3 while the deployed membranes are illustrated in
Figure 4.

Table 1: Material Properties of Membrane and Spiral Spring

Parameter Kapton Steel
Elasticity Modulus 2.5 GPa 200 GPa
Poisson Ratio 0.34 0.29
Density 1420 kg/m3 7800 kg/m3

The starting point of the deployment is the current produced by on-board computer (OBC)
command. Lock mechanism consists of dynamo wire and it is cut by the help of passing current.
Then, the trigger bands become unlocked and transmit the required deployment energy from spiral
springs to storage cylinders. The storage cylinders provide membrane deployment by rotating. OBC

The 4S Symposium 2012 – P. Acar, M. Nikbay and A. R. Aslan 2

command is issued at the end of satellite mission while de-orbiting process is expected to last
within 25 years after deployment.

Figure 1. Subsystems of 3USAT

Figure 2. General Schematic of 3USAT De-Orbiting Mechanism [11]

The 4S Symposium 2012 – P. Acar, M. Nikbay and A. R. Aslan 3

 Figure 3. Lock Mechanism of 3USAT De-Orbiting Mechanism [11]

The main principle of de-orbiting is aerodynamic drag enhancement. General formulation of

aerodynamic drag force is given in (1).

1
D V 2 ScD (1)
2

where V is orbital tangential velocity of the satellite,  is atmospheric density. Aerodynamic drag
coefficient, cD , is assumed to be equal to 2.2 for LEO satellites [12]. S indicates effective surface
area, which is directly proportional to membrane dimensions and orientation of the satellite.

The drag force is calculated by considering the worst-case conditions which involve maximum
satellite altitude for de-orbiting, 680 km, where the atmospheric density has its minimum value. The
specified altitude also corresponds to the minimum value of orbital tangential velocity of the
satellite, whose formulation is defined in (2).

G ME
V (2)
RE  h

where G  6.67428 1011 Nm2 /kg2 is universal constant of gravitation and h indicates the satellite
altitude. M E  5.97214 1024 kg and RE  6.37814 106 m denote the mass and radius of the Earth
respectively.

The main concern of the present work is to provide an orbit decay which ends up with a final
altitude of 180 km. The initial altitude of 3USAT, where the de-orbiting mechanism is deployed, is
specified as 680 km. Thus, the major aim is to decay the satellite altitude from 680 km to 180 km.

The 4S Symposium 2012 – P. Acar, M. Nikbay and A. R. Aslan 4

 Figure 4. Deployable Membranes [11]

The de-orbiting system is deployed by the mechanical energy of spiral springs. Moment generated
by a spiral spring is defined in (3) while the stress due to acted forces on the spring is given in (4).
 Ebs ts3
M (3)
6L

6M
S (4)
bs ts2

where E is elasticity modulus of steel, bs and t s are width and thickness of spiral spring
respectively, and n denotes number of spins.

The maximum available pressure is regarded for stress calculations in spiral spring. Moreover, the
minimum aerodynamic drag force corresponding to the maximum altitude of the satellite is
considered for maximum system reliability. An in-house code developed for the calculation of
minimum aerodynamic drag force and spring moment is coupled with the optimization software,
modeFRONTIER to perform multi-objective design optimization of de-orbiting mechanism.

3 MULTI-OBJECTIVE DESIGN OPTIMIZATION OF 3USAT DE-ORBITING SYSTEM

The main goal of the present work is to design an efficient de-orbiting system providing the shortest
de-orbiting time with an easily deployed mechanism. Therefore, maximization of aerodynamic drag
force is acquired as a design objective so as to reduce the orbital life time. Furthermore,
maximization of spiral spring moment is determined as the second objective to enable an easy and
safe deployment. Maximization of drag force and spring moment can cause extreme increase in
weight which can break the structural integrity. Thus, a design constraint is defined to restrict the
mass of the deployable mechanism. Stress due to acted forces on the spiral spring requires the
definition of another constraint since the spiral spring must be able to work without any structural
failure. The optimization variables are ascertained as length and width of membrane structure,
thickness and width of spiral spring. The mathematical form of optimization problem is given in (5)

The 4S Symposium 2012 – P. Acar, M. Nikbay and A. R. Aslan 5

to (9).

max D(s), max M (s) (5)

m( s)
g1 ( s)  1  0 (6)
mmax

S ( s)
g 2 ( s)  1  0 (7)
Smax

S  s  sL  s  sU  (8)

s  Lm , Bm , d s , ts , bs  (9)

where g1 ( s) and g2 ( s) denote the constraints defined for maximum allowable mass and stress
respectively while s shows optimization variables. sL and sU indicate lower and upper limits of
optimization parameters.

The multi-objective design optimization is performed by the coupling of the in-house code and
optimization software, modeFRONTIER. The optimization workflow shown in Figure 5, basically
illustrates the input and output variables, design objectives and constraints, MATLAB coupling and
the optimization algorithm. Lower and upper limits of the optimization parameters are given in
Table 2.

Figure 5. Workflow for Multi-Objective Optimization Problem

The 4S Symposium 2012 – P. Acar, M. Nikbay and A. R. Aslan 6

NSGA-II (Multi-Objective Genetic Algorithm-II) is preferred as optimization algorithm since
genetic algorithms are meta-heuristic, thus they are particularly well-suited for this class of
problems [13].
Table 2: Lower and Upper Limits of Design Variables
Parameter sL sU
Lm 0.2 m 3.0 m
Bm 40 mm 70 mm
ds 10 mm 50 mm
ts 0.1 mm 0.2 mm
bs 10 mm 20 mm

The optimization took about 2 hours 57 minutes on a platform which has Intel Core 2 Duo CPU
T6500@2.10GHz Processor and 4GB RAM on Microsoft Windows 7 32 Bit Operating System.
13899 feasible and 6101 infeasible designs are acquired for 200 DoE (Design of Experiments) with
Sobol Sequence. Since multi-objective design optimization is performed, pareto optimal solutions
exist. The pareto optimal solutions are listed in Table 3 by considering design objectives and
constraints.

Table 3: Pareto Optimal Solutions of Design Optimization

Pareto Design D (kN) M (kN.m) m (kg) S (MPa)
Pareto-1 6.3714 1599.9951 0.2 1199.9963
Pareto-2 6.2944 1599.9989 0.1976 1199.9993
Pareto-3 6.1472 1599.9990 0.1930 1199.9993
Pareto-4 6.3688 1599.9988 0.2 1199.9991

All pareto designs given in Table 3 have similar constraint values. Therefore, the design which
produces the maximum drag force, Pareto-1, is selected as optimum design since spiral spring
moment values are even very similar. The parameters of optimum design are shown in Table 4.

Table 4: Design Parameters of Optimum Design

Lm Bm ds bs ts
2.9351 m 69.8876 mm 34.50 mm 20 mm 0.2 mm

The de-orbiting time of the optimum design will be calculated in the next section.

4 CALCULATION OF DE-ORBITING TIME

De-orbiting time of the optimum configuration is calculated in STK by considering J2 perturbation

propagator and NRLMSISE-00 atmospheric model. The cross-sectional drag area of the optimum
configuration is taken into account in addition to the surface area of 3USAT. The de-orbiting time is
determined as 16.8 years. The orbital life-time of the satellite satisfies IADC’s suggestion while
still providing an efficient and easily deployed mechanism. Variations on perigee and apogee
distances, and eccentricity during orbit decay process are illustrated in Figure 6.

The 4S Symposium 2012 – P. Acar, M. Nikbay and A. R. Aslan 7

5 CONCLUSIONS

Multi-objective design optimization of a 3U satellite is performed to achieve an efficient and easily

deployed de-orbiting system. The de-orbiting mechanism of interest is being developed by ITU-
SSDTL within the context of 3USAT satellite project. The de-orbiting system is a mechanically
deployable structure while the working principle is based on aerodynamic drag enhancement.

Figure 6. Progression of Orbit Decay

The expectation of the present work is to provide the shortest time de-orbiting with a structurally
safe mechanism. Therefore, the aerodynamic drag force and spring moment are maximized by
introducing design parameters which are dimensions of deployable membrane structure and spiral
spring. Design constraints for maximum allowable system mass and stress acted on spiral spring are
defined to maintain structural efficiency while still consuming less time for de-orbiting. A
MATLAB code is developed to set up all design criteria and parameters, and coupled with the
optimization software.

The pareto optimal designs generated by multi-objective design optimization are evaluated whether
they can provide considerable improvement in drag force while still producing reasonable spring
moment. The optimal design is selected by mostly taking cognizance of aerodynamic drag force.
Finally, the de-orbiting time of the optimum design is calculated with STK. The final design highly
satisfies IADC’s de-orbiting time suggestion.

The future work may contain design of such mechanisms based on robustness criterion since the
variations in space environment require the most reliable systems. Reliability and robustness based
applications can be extended for all satellite components instead of a subsystem-level consideration.

The 4S Symposium 2012 – P. Acar, M. Nikbay and A. R. Aslan 8

 6 ACKNOWLEDGEMENTS

The contributions of Ceyhun Tola to the design of 3USAT de-orbiting mechanism are highly
acknowledged. The first author would also like to thank to Doğa Can Su Öztürk for her support in
many difficult times.

7 REFERENCES

[1] Liou J. C., Johnson N. L., Risk in Space from Orbiting Debris, Science, Vol. 331, No. 5759, pp:
256-269, 2006.

[2] IADC Space Debris Mitigation Guidelines, Inter-Agency Space Debris Coordination Committee
Document: IADC-02-01, 2002.

[3] Forward R. L., Hoyt R. P., Uphoff, C. W., Terminator TetherTM: A Spacecraft De-Orbit
Device, Journal of Spacecraft and Rockets, Vol. 37, No. 2, pp. 187-196, 2000.

[4] Bombardelli C., Nuez-Aylln J. M., Pelez J., Performance Analysis of Bare Electrodynamic
Tethers as Microsat Deorbiting Systems, 20th AAS/AIAA Space Flight Mechanics Meeting, San
Diego, USA, 2010.

[5] Corsi J., Iess L., Stability and Control of Electrodynamic Tethers for De-Orbiting Applications,
Acta Astonautica, Vol. 48, No. 5-12, pp. 491-501, 2001.

[6] Lappas V., Adeli N., Visagie L., Fernandez, J., Theodorou, T., Steyn, W., Perren, M., Cubesail:
A Low Cost CubeSat Based Solar Sail Demonstration Mission, Advances in Space Research, Vol.
48, No. 11, pp. 1890-1901, 2011.

[7] Andrews J., Watry K., Brown K., Nanosat Deorbit and Recovery System to Enable New
Missions, 25th Annual AIAA/USU Conference on Small Satellites, Logan, Utah, USA, 2011.

[8] Janhunen P., Electrostatic Plasma Brake for Deorbiting a Satellite, Journal of Propulsion and
Power, Vol. 26, No. 2, pp. 370-372, 2010.

[9] Maessen D. C., van Breukelen E., Zandbergen B., Bergsma O., Development of a Generic
Inflatable De-Orbit Device for CubeSats, 58th International Astonautical Congress, Hyderabad,
India, 2007.

[10] Lokcu E., Ash R. L., A De-Orbit System Design for CubeSat Payloads, 5th International
Conference on Recent Advances in Space Technologies (RAST), Istanbul, Turkey, 2011.

[11] Tola C., Aslan A. R., Ilarslan, M., Development of a De-orbiting System for a 3-Unit CubeSat,
3rd Nano-Satellite Symposium, Japan, 2011.

[12] Storch J. A., Aerodynamic Disturbances On Spacecraft In Free-Molecular Flow, Aerospace

Report No. TR-2003 (3397)-1, The Aerospace Corporation, EI, Segundo, CA, 2003.

[13] Konak A., Coit D. W., Smith A. E, Multi-Objective Optimization Using Genetic Algorithms: A
Tutorial, Reliability Engineering and System Safety, Vol. 91, No. 9, pp.992-1007, 2006.

The 4S Symposium 2012 – P. Acar, M. Nikbay and A. R. Aslan 9

View publication stats