Neutron Acceptance Diagram Shading Phil Bentley ILL Monday, 17 - - PowerPoint PPT Presentation

▶
neutron acceptance diagram shading
SMART_READER_LITE
LIVE PREVIEW

Neutron Acceptance Diagram Shading Phil Bentley ILL Monday, 17 - - PowerPoint PPT Presentation

May 22, 2023 444 likes •974 views

Neutron Acceptance Diagram Shading Phil Bentley ILL Monday, 17 August 2009 Why? New problems in optics Nature has solved such often involve optimisation problems optimisation in many- already dimensional parameter spaces May

slide-1
SLIDE 1

Neutron Acceptance Diagram Shading

Phil Bentley ILL

Monday, 17 August 2009

slide-2
SLIDE 2

Why?

  • New problems in optics
  • ften involve
  • ptimisation in many-

dimensional parameter spaces

  • May have lots of local
  • ptima
  • Analytical solutions

might not exist

  • Nature has solved such
  • ptimisation problems

already

Monday, 17 August 2009

slide-3
SLIDE 3

Why?

  • Particle swarms / genetic

algorithms need 100 iterations or more to converge

  • Each iteration evaluates

30-50 agents

  • Therefore, 3000

simulations minimum

  • Either use a cluster or

something other than Monte-Carlo

Monday, 17 August 2009

slide-4
SLIDE 4
  • Complex geometry with

many coupled parameters

  • SPAN-cryopad
  • multi-channel guides IN5
  • WASP
  • TYREX 3He polariser

(new model)

A.I. Works Well

Monday, 17 August 2009

slide-5
SLIDE 5
  • Complex geometry with

many coupled parameters

  • SPAN-cryopad
  • multi-channel guides IN5
  • WASP
  • TYREX 3He polariser

(new model)

A.I. Works Well

Monday, 17 August 2009

slide-6
SLIDE 6
  • Complex geometry with

many coupled parameters

  • SPAN-cryopad
  • multi-channel guides IN5
  • WASP
  • TYREX 3He polariser

(new model)

A.I. Works Well

Monday, 17 August 2009

slide-7
SLIDE 7
  • Complex geometry with

many coupled parameters

  • SPAN-cryopad
  • multi-channel guides IN5
  • WASP
  • TYREX 3He polariser

(new model)

A.I. Works Well

Monday, 17 August 2009

slide-8
SLIDE 8

Full Optimisations

  • Simulate the instrument

all the way from moderator to detector

  • All parameters free
  • Don’t care if parameters

are strongly coupled

  • Parameter limit: >47 but

<650 with canonical algorithms

Monday, 17 August 2009

slide-9
SLIDE 9

SANS on D22

  • 100 000 000 trajectories entering the guide
  • 666 seconds CPU time (11 minutes) => 74

trajectories hit the sample

  • Statistical error = 12%
  • 1% error needs 25 hours
  • Optimisation of D22: 8-25 years (1 cpu)

Monday, 17 August 2009

slide-10
SLIDE 10

SANS on D22

  • 100 000 000 trajectories entering the guide
  • 666 seconds CPU time (11 minutes) => 74

trajectories hit the sample

  • Statistical error = 12%
  • 1% error needs 25 hours
  • Optimisation of D22: 8-25 years (1 cpu)

Monday, 17 August 2009

slide-11
SLIDE 11

SANS on D22

  • 100 000 000 trajectories entering the guide
  • 666 seconds CPU time (11 minutes) => 74

trajectories hit the sample

  • Statistical error = 12%
  • 1% error needs 25 hours
  • Optimisation of D22: 8-25 years (1 cpu)

< 1 minute

Monday, 17 August 2009

slide-12
SLIDE 12

SANS on D22

  • 100 000 000 trajectories entering the guide
  • 666 seconds CPU time (11 minutes) => 74

trajectories hit the sample

  • Statistical error = 12%
  • 1% error needs 25 hours
  • Optimisation of D22: 8-25 years (1 cpu)

< 1 minute < ~1 week

Monday, 17 August 2009

slide-13
SLIDE 13

nads Speed Gain

1.5 x

http://flickr.com/photos/17393884@N00/5341048/ Flickr http://www.apolloarchive.com/apollo_gallery.html

Monday, 17 August 2009

slide-14
SLIDE 14

Similar Problem Solved

  • Ray tracing is routinely

used in movies

  • Very slow! Scenes

rendered on huge clusters

  • Vector graphics is close

to ray trace quality

  • 30 frames per second

Monday, 17 August 2009

slide-15
SLIDE 15

Similar Problem Solved

  • Ray tracing is routinely

used in movies

  • Very slow! Scenes

rendered on huge clusters

  • Vector graphics is close

to ray trace quality

  • 30 frames per second

Monday, 17 August 2009

slide-16
SLIDE 16

Neutron Bunches

  • The analogy is to group trajectories into

“similar” bunches and treat the group as

  • ne single object.
  • Bunches are phase space regions with

linear relation between trajectories

  • How do we separate the bunches so the

calculation remains accurate?

Monday, 17 August 2009

slide-17
SLIDE 17

Guide Reflectivity

  • Divide bunches along

m=1 and critical m.

  • Trajectories within a

bunch are linearly dependent

  • Guide reflectivity is an

idealised curve

  • R at m_crit taken from

Swiss neutronics

Monday, 17 August 2009

slide-18
SLIDE 18

Acceptance Diagrams

  • Define boundaries in distance-divergence

space

  • Division is on module-by-module basis

Monday, 17 August 2009

slide-19
SLIDE 19

Source Module

  • Just like Monte-Carlo,

we have a source plane and a virtual “exit window”.

  • Exit window must be at

least as large as subsequent module

  • Propagation = shear

Monday, 17 August 2009

slide-20
SLIDE 20

Source Module

  • Just like Monte-Carlo,

we have a source plane and a virtual “exit window”.

  • Exit window must be at

least as large as subsequent module

  • Propagation = shear

Monday, 17 August 2009

slide-21
SLIDE 21

Source Module

  • Just like Monte-Carlo,

we have a source plane and a virtual “exit window”.

  • Exit window must be at

least as large as subsequent module

  • Propagation = shear

Monday, 17 August 2009

slide-22
SLIDE 22

Collimator/Aperture

  • For aperture clip off

extrema in space axis

  • For collimator, clip off

extrema in divergence axis

Monday, 17 August 2009

slide-23
SLIDE 23

Collimator/Aperture

  • For aperture clip off

extrema in space axis

  • For collimator, clip off

extrema in divergence axis

Monday, 17 August 2009

slide-24
SLIDE 24

Collimator/Aperture

  • For aperture clip off

extrema in space axis

  • For collimator, clip off

extrema in divergence axis

Monday, 17 August 2009

slide-25
SLIDE 25

Change Coordinates

  • Rotate beam axis
  • Translate beam axis
  • Both are a translation of

phase space volumes

Monday, 17 August 2009

slide-26
SLIDE 26

Guide Module

  • Kill neutrons that miss

the entrance

  • Propagate neutrons to

the end of the guide

  • Divide along m

boundaries

  • Multiply reflected weight

by reflectivity

Monday, 17 August 2009

slide-27
SLIDE 27

Guide Module

  • Kill neutrons that miss

the entrance

  • Propagate neutrons to

the end of the guide

  • Divide along m

boundaries

  • Multiply reflected weight

by reflectivity

Monday, 17 August 2009

slide-28
SLIDE 28

Guide Module

  • Kill neutrons that miss

the entrance

  • Propagate neutrons to

the end of the guide

  • Divide along m

boundaries

  • Multiply reflected weight

by reflectivity

Monday, 17 August 2009

slide-29
SLIDE 29

Statistical Weight

  • Use triangular primitives
  • Statistical weight of a

triangle is a volume calculation

  • Right-wedge plus

square-based pyramid

  • Only difficulty is very

thin triangles (known problem in CS)

Monday, 17 August 2009

slide-30
SLIDE 30

Guides Supported

  • Straight
  • Curved parallel
  • Converging
  • Diverging
  • Arbitrary

Monday, 17 August 2009

slide-31
SLIDE 31

Test 1: Simple Guide

Monday, 17 August 2009

slide-32
SLIDE 32

Test 1: Simple Guide

Monday, 17 August 2009

slide-33
SLIDE 33

Test 1: Simple Guide

Monday, 17 August 2009

slide-34
SLIDE 34

Test 1: Simple Guide

Monday, 17 August 2009

slide-35
SLIDE 35

Test 1: Simple Guide

2% Error, 4.8 s 48 ms

Monday, 17 August 2009

slide-36
SLIDE 36

Test 2: D22 (V)

Monday, 17 August 2009

slide-37
SLIDE 37

Test 2: D22 (V)

Monday, 17 August 2009

slide-38
SLIDE 38

Test 2: D22 (V)

Monday, 17 August 2009

slide-39
SLIDE 39

Test 2: D22 (H)

Monday, 17 August 2009

slide-40
SLIDE 40

Test 2: D22 (H)

Monday, 17 August 2009

slide-41
SLIDE 41

Test 2: D22 (H)

Monday, 17 August 2009

slide-42
SLIDE 42

Test 2: D22 (H)

Monday, 17 August 2009

slide-43
SLIDE 43

Test 2: D22 (H)

Calculate flux on sample: 55 ms

Monday, 17 August 2009

slide-44
SLIDE 44

Test 3: WASP

  • Agreement between

NADS and MCSTAS on sample

  • MCSTAS model

Independently calculated by Peter Fouquet from moderator to Sample

  • ~5 seconds for nads per

wavelength (3 mins for white beam)

Monday, 17 August 2009

slide-45
SLIDE 45

Test 3: WASP

  • Agreement between

NADS and MCSTAS on sample

  • MCSTAS model

Independently calculated by Peter Fouquet from moderator to Sample

  • ~5 seconds for nads per

wavelength (3 mins for white beam)

MCSTAS nads

Monday, 17 August 2009

slide-46
SLIDE 46

Test 4: Thales

  • Agreement between

SIMRES / RESTRAX and nads

  • Trace up to virtual

source (M3)

  • ~40 ms for nads per

wavelength (1.7 sec for white beam)

4

ThALES

M2 M1 M3 M4

.

Monday, 17 August 2009

slide-47
SLIDE 47

Test 4: Thales

  • Agreement between

SIMRES / RESTRAX and nads

  • Trace up to virtual

source (M3)

  • ~40 ms for nads per

wavelength (1.7 sec for white beam)

4

ThALES

M2 M1 M3 M4

.

SIMRES nads

Monday, 17 August 2009

slide-48
SLIDE 48

For Coders

  • nads kernel in C++
  • Understands xml input

from file or pipe

  • xml schema lets you

check syntax of hand- coded simulations

  • Uses getopt with varying

degrees of verbosity

  • nads -q prints only the
  • utput flux
  • Fully scriptable using

bash / c / mathematica...

Monday, 17 August 2009

slide-49
SLIDE 49
  • GUI
  • 3d instrument visualisation with OpenGL
  • Mathematica scripts

Live Demonstration

Monday, 17 August 2009

slide-50
SLIDE 50

Limitations

  • Each nads calculation is monochromatic

(white beam possible via looping)

  • Small angle approximation
  • Separable vertical & horizontal channels
  • Round objects approximated by rectangular
  • bjects of equal area (round sources etc)

Monday, 17 August 2009

slide-51
SLIDE 51

Acknowledgements

  • ESFRI
  • Ken Andersen
  • Klaus Habicht (HMI / BLZ)
  • Leo Cussen (Cussen Consulting)

Monday, 17 August 2009