P. Ben-Abdallah , J.M. Llorens , L. Bergamin , and T. · PDF fileADVANCED CONCEPTS TEAM 23 4. Conclusions and Outlook Conclusions •Development of a Reversal Engineering Tool for

ADVANCED CONCEPTS TEAM 1

ADVANCED CONCEPTS TEAM

P. Ben-Abdallah1, J.M. Llorens2, L. Bergamin2, and T. Vinko2

Esa 6th Round Table on MNTs

e

x

1 CNRS, Ecole Polytechnique, Laboratorie de Thermocinétique, Nantes, France2 Advanced Concepts Team, ESA (ESTEC), Noordwijk, The Netherlands


ACT AssessmentUniversities

• Informatics• Mission Analysis• Nanotechnologies• Propulsion

What is the ACT?

Who constitute the ACT?

How the ACT works?

• Loose ideas linked to space• First basic filtering of loose ideas linked to space• Preliminary ACT assessments - second filter• Academic assessment ARIADNA• Eventual transfer to ESA R&D programmes

Loose ideas linked to space

D/TEC - Programmes - GSP

filte

red,

ab

ando

ned

idea

s

filtered,

abandoned

ideas

Ariadna

Multidisciplinary research group:•Artificial Intelligence•Biomimetics•Energy Systems•Fundamental Physics

• 2 Staff • 6 Research Fellows• 5 Young Graduate Trainee


1.Introduction to directional

microstructured radiators

2.Reverse engineering strategy

3.Case studies:

•Quasi-isotropic radiator at room

temperature

•Directional radiator in the near

infrared

4.Conclusions and Outlook


1. Introduction to directionalmicrostructured radiators




3.Case studies:


temperature


infrared




General Problem: Thermal control of Spacecraft

Solarradiation

Conduction

Convection

Radiation

XX

Proposal: Design surface finishes withspatially coherent thermal radiation

1.Polar materials + surface grating

2.Photonic band gap materials

3.Polar materials / metallic layer + photonic structures

4.Left-Handed material



General Problem: Thermal control of Spacecraft

Solarradiation

Conduction

Convection

Radiation

XX

Proposal: Design surface finishes withspatially coherent thermal radiation

1.Polar materials + surface grating

2.Photonic band gap materials

3.Polar materials / metallic layer + photonic structures

4.Left-Handed material



Polar material (SiC, GaN,…)

High and Low refractive indexPhotonicstructure

Schematic description of the concept:

Metalic layer (Ag, Al,…)ε(ω)

nH,L

Directionalradiation

Stop-band

λ

r

ReflectivityTransmitivityAbsorptivityEmissivity

t

r

eα

e

x

s, por

TE, TM


2. Reverse engineering strategy




3.Case studies:


temperature


infrared



Objective of the study: For a given radiation distribution find the suitablemultilayer structure.

InputOptimization

process

Outputn(λ, z)

Optimization solver: Genetic Algorithm

Multilayer codification

Array element

Material layer

Initial random population

Evolution with genetic operators

Best solution

e(λ, θ) r(λ, θ)


Fitness function:

Physical model: Transfer matrix formalism

E+

E− E0−

E0+E0−

E0+E+

E−Transfermatrix

Reflection + Transmission

Energy conservation + Kirchhoff’s law

Emission

J =

Z λmax

λmin

Z θmax

θmin

(es − eTarget)2 + (ep − eTarget)2

+(rs − rTarget)2 + (rp − rTarget)2 dλ dθ


Implementation of the GA algorithm

Single CPU Distributed Computation (DC)

Single CPU evolves all the population

Each CPU evolves apart of the population

Physical DC platform:

Server

Client

Client

ClientClient

Population Initialization

+Sub-population

distribution

Population evolution


3. Case studies




3.Case studies:


temperature


infrared



3. Case studies

Quasi-isotropic radiator at room temperature

λmax ≈ 10 μm

Radiation distribution T=300 K

Acording to Wien´s law:

•Polar Material: 3C-SiC (emission at 12.6 μm)•Low index material: CdTe nL=1.6•High index material: Ge nH=4.0

θ

Combination of materials

Parameters of the structure•50 layers => 350 (~7x1025) possible combinations•100 nm thickness


3. Case studies


Target Function

λ(μm

)

0 2 4 6 8 10

4

5

6

7

8

9

J (x

10-2

)Function Evaluantion (x104)

N. Generations ≡ 100 Function Eval.

Ge 3C-SiC CdTe

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 10 20 30 40 50 60 70 80 90 12.2

12.3

12.4

12.5

12.6

12.7

12.8

θJ

(x 1

0-3 )


3. Case studies

Quasi-isotropic radiator at room temperatureE

mis

sion

0 10 20 30 40 50 60 70 80 12.2

12.3

12.4

12.5

12.6

12.7

12.8

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 10 20 30 40 50 60 70 80 12.2

12.3

12.4

12.5

12.6

12.7

12.8

Ref

lect

ion

0 10 20 30 40 50 60 70 80 12.2

12.3

12.4

12.5

12.6

12.7

12.8

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 10 20 30 40 50 60 70 80 12.2

12.3

12.4

12.5

12.6

12.7

12.8

s

s

p

p

λ

λ

λ

λ

θ

θ

θ

θ


3. Case studies


mis

sion

Ref

lect

ion

s

s

p

p

λ

λ

λ

λ

θ

θ

θ

θ

0 10 20 30 40 50 60 70 80 90 8

9

10

11

12

13

14

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 10 20 30 40 50 60 70 80 90 8

9

10

11

12

13

14

0 10 20 30 40 50 60 70 80 90 8

9

10

11

12

13

14

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 10 20 30 40 50 60 70 80 90 8

9

10

11

12

13

14


3. Case studies


Run Details:1. Single CPU

• ESAGRID (Pentium Xeon2.40 GHz)

• Average functionevaluation time: ~ 2 s

2. DC• 12 Dell Desktops• Average function

evaluation time: ~ 5 s0 2 4 6 8 10 12

4

5

6

7

8

9

J (x

10-2

)

Function Evaluantion (x104)

SingleDC

Implementation into the DCJ

(x 1

0-3 )

6,700121,500

Outperformance ~ factor 18


3

4

5

6

7

8

9

100 1000 10000 100000

3. Case studies


SingleDC

Time reduction

J (x

10-

3 )

Time (s)

T=235,000 (2 d 17 h)

T=33,500 (9 h)

Outperformance ~ factor 7


3. Case studies

Directional radiator in the near infrared

Radiation distribution Working spectral range

[1.8 μm – 2.8 μm]

•Metal Layer: Ag (broad emission)•Low index material: SiO2 nL=1.45•High index material: Si nH=3.3

Combination of materials

Parameters of the structure•50 layers => 350 (~7·1025) possible combinations•50 nm thickness

e

x


0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 10 20 30 40 50 60 70 80 90 1.8

1.9

2

2.1

2.2

2.3

2.4

2.5

2.6

2.7

2.8

3. Case studies


Target Function

θ

λ(μm

)

N. Generations ≡ 100 Function Eval.

0 2 4 6 8 103.75

4.00

4.25

4.50

4.75

5.00

J (x

10-2

)Function Evaluantion (x104)

Si Ag SiO2

J (x

10-

3 )


3. Case studies


mis

sion

Ref

lect

ion

s

s

p

p

λ

λ

λ

λ

θ

θ

θ

θ

0 10 20 30 40 50 60 70 80 90 1.8

1.9

2

2.1

2.2

2.3

2.4

2.5

2.6

2.7

2.8

0 10 20 30 40 50 60 70 80 90 1.8

1.9

2

2.1

2.2

2.3

2.4

2.5

2.6

2.7

2.8

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 10 20 30 40 50 60 70 80 90 1.8

1.9

2

2.1

2.2

2.3

2.4

2.5

2.6

2.7

2.8

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 10 20 30 40 50 60 70 80 90 1.8

1.9

2

2.1

2.2

2.3

2.4

2.5

2.6

2.7

2.8


4. Conclusions and Outlook




3.Case studies:


temperature


infrared



4. Conclusions and Outlook

Conclusions

•Development of a Reversal Engineering Tool for designing directional microstructured radiators •Proof their suitability for different types of radiators•Implementation of a distributed version of the GA solver•Distributed strategy shows an outperformance of factor 7

Outlook

• Asses the feasibility of the optimized structure (explore materials for the active layer)• Study the “control” over the reflectivity and emissivity features• Implement other GO solvers into the tool (DIGMO) • Combine this proposal with surface gratings to improve the directivity


Thank you for your attention !

More Information:• About the ACT: https://www.esa.int/act• About ARIADNA: http://www.esa.int/gsp/ACT/ariadna/index.htm• About Microstructured Radiators: Ariadna Final Report

http://www.esa.int/gsp/ACT/doc/ARI/ARI%20Study%20Report/ACT-RPT-NAN-ARI-069501-Micro_Radiators-Nantes.pdf

• About DIGMO: http://www.esa.int/gsp/ACT/inf/op/act-dc_digmo.htm

Documents

P. Ben-Abdallah , J.M. Llorens , L. Bergamin , and T. · PDF fileADVANCED CONCEPTS TEAM 23 4. Conclusions and Outlook Conclusions •Development of a Reversal Engineering Tool for