Top Banner
Towards an improved PEPT triangulation routine J Newling 1 , AJ Morrison 1 , N Fowkes 2 , I Govender 1 and L Bbosa 1 1 University of Cape Town, Cape Town, South Africa 2 University of Western Australia, Perth, Australia
19

Towards an improved PEPT triangulation routine J Newling 1, AJ Morrison 1, N Fowkes 2, I Govender 1 and L Bbosa 1 1 University of Cape Town, Cape Town,

Dec 17, 2015

Download

Documents

Randall Fowler
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Towards an improved PEPT triangulation routine J Newling 1, AJ Morrison 1, N Fowkes 2, I Govender 1 and L Bbosa 1 1 University of Cape Town, Cape Town,

Towards an improved PEPT triangulation routine

J Newling1, AJ Morrison1, N Fowkes2, I Govender1 and L Bbosa1

 1 University of Cape Town, Cape Town, South Africa

2University of Western Australia, Perth, Australia

Page 2: Towards an improved PEPT triangulation routine J Newling 1, AJ Morrison 1, N Fowkes 2, I Govender 1 and L Bbosa 1 1 University of Cape Town, Cape Town,

Tumbling Mills

• Minerals industry (gold, platinum, copper, etc …)

• Main aim is size reduction of extracted ore

• Very energy-intensive, however inefficient

• Aggressive environment,in situ measurement not feasible

• Models are empirical– Mill specific– Ore specificMill diameter 0.3 – 5m

Rotational speed 15 – 40 rpm

Page 3: Towards an improved PEPT triangulation routine J Newling 1, AJ Morrison 1, N Fowkes 2, I Govender 1 and L Bbosa 1 1 University of Cape Town, Cape Town,

Positron Emission Particle Tracking

Page 4: Towards an improved PEPT triangulation routine J Newling 1, AJ Morrison 1, N Fowkes 2, I Govender 1 and L Bbosa 1 1 University of Cape Town, Cape Town,

γ

γ

Positron Emission Particle Tracking

Page 5: Towards an improved PEPT triangulation routine J Newling 1, AJ Morrison 1, N Fowkes 2, I Govender 1 and L Bbosa 1 1 University of Cape Town, Cape Town,

Positron Emission Particle Tracking

Page 6: Towards an improved PEPT triangulation routine J Newling 1, AJ Morrison 1, N Fowkes 2, I Govender 1 and L Bbosa 1 1 University of Cape Town, Cape Town,

Sources of false events

      True Pairing

        Scattered Pairing

       Random Pairing

Page 7: Towards an improved PEPT triangulation routine J Newling 1, AJ Morrison 1, N Fowkes 2, I Govender 1 and L Bbosa 1 1 University of Cape Town, Cape Town,

Triangulation

100200

300400

500

0

200

400

6000

100

200

300

400

500

600

75% - 90% of recorded events are discarded

Page 8: Towards an improved PEPT triangulation routine J Newling 1, AJ Morrison 1, N Fowkes 2, I Govender 1 and L Bbosa 1 1 University of Cape Town, Cape Town,

Proposal 1: Minimum perpendicular distance method

Method•Find the midpoint of the perpendicular between successive lines of response•Use the median of these midpoints to estimate the particle location in that time interval 

Motivation•Avoid iteration by using the median to weight true pairs

Shortcoming•No guarantee that the closest approach is in the area of the tracer particle

Page 9: Towards an improved PEPT triangulation routine J Newling 1, AJ Morrison 1, N Fowkes 2, I Govender 1 and L Bbosa 1 1 University of Cape Town, Cape Town,

Proposal 1: Minimum perpendicular distance method

Page 10: Towards an improved PEPT triangulation routine J Newling 1, AJ Morrison 1, N Fowkes 2, I Govender 1 and L Bbosa 1 1 University of Cape Town, Cape Town,

Proposal 2: Density of lines

Method•Discretise the field of the view into a 3D grid.•Use the number of intersections of the LoRs with each grid element to isolate the particle position

Motivation•Discriminate against random and scattered events

Shortcoming•Computationally expensive

Page 11: Towards an improved PEPT triangulation routine J Newling 1, AJ Morrison 1, N Fowkes 2, I Govender 1 and L Bbosa 1 1 University of Cape Town, Cape Town,

Proposal 2: Density of linesDino Giovannoni & Matthew Bickell (Physics Honours)

From detected lines to line density…

… to probability distributions…

… to particle position.

Page 12: Towards an improved PEPT triangulation routine J Newling 1, AJ Morrison 1, N Fowkes 2, I Govender 1 and L Bbosa 1 1 University of Cape Town, Cape Town,

Proposal 3: 2D triangulation

Method•Divide LoR into coplanar sets and use these to reduce the problem to a 2D one

Motivation•Simplify the 3D case into a 2D problem

Shortcoming•Drastically reduces the statistics•Does not discriminate between true and false lines.

Page 13: Towards an improved PEPT triangulation routine J Newling 1, AJ Morrison 1, N Fowkes 2, I Govender 1 and L Bbosa 1 1 University of Cape Town, Cape Town,

Proposal 4: Distance distribution 

Method•Use the current iterative method to calculate the centroid•Use the distribution of LoR distances from the centroid to dynamically determine the fraction to discard•Recalculate the centroid and repeat until some convergence criteria is met.

Motivation•Avoid having to calibrate the routine for each experiment

Shortcoming•Does not reduce the computational expense

Page 14: Towards an improved PEPT triangulation routine J Newling 1, AJ Morrison 1, N Fowkes 2, I Govender 1 and L Bbosa 1 1 University of Cape Town, Cape Town,

Proposal 4: Distance distribution 

Frequency of events

Distance from centroid /mm

Page 15: Towards an improved PEPT triangulation routine J Newling 1, AJ Morrison 1, N Fowkes 2, I Govender 1 and L Bbosa 1 1 University of Cape Town, Cape Town,
Page 16: Towards an improved PEPT triangulation routine J Newling 1, AJ Morrison 1, N Fowkes 2, I Govender 1 and L Bbosa 1 1 University of Cape Town, Cape Town,
Page 17: Towards an improved PEPT triangulation routine J Newling 1, AJ Morrison 1, N Fowkes 2, I Govender 1 and L Bbosa 1 1 University of Cape Town, Cape Town,
Page 18: Towards an improved PEPT triangulation routine J Newling 1, AJ Morrison 1, N Fowkes 2, I Govender 1 and L Bbosa 1 1 University of Cape Town, Cape Town,
Page 19: Towards an improved PEPT triangulation routine J Newling 1, AJ Morrison 1, N Fowkes 2, I Govender 1 and L Bbosa 1 1 University of Cape Town, Cape Town,

Conclusion