High spatial and temporal

resolution planar

capacitive sensing in oil-

water pipe flow

ADRIANA BONILLA RIAÑO

ADVISOR: DR. ANTONIO BANNWART

CO- ADVISOR: DR. OSCAR M. H. RODRIGUEZ

NOVEMBER 29, 2013

1

OUTLINE

• Holdup Estimation in Core Flow Using Image

Processing

• Film thickness measurement using infrared

images

• Planar sensor simulation

• Next steps

2

Holdup estimation in core flow using image processing

Oil-water core flow in a 62mm i.d. pipe

Heavy oil (2000 cP and 950 kg/m3 at 35 °C) and water.

4 experiments with 100 images each.

3

Algorithm

4

Pre-processing

• Cutting

• Noise Reduction

• Contrast enhancement

Algorithm

5

Segmentation

• Active

contours

• Mathematical

morphology

Algorithm

6

Volume Calculation

• Radius measurement

• Polynomial fit

• Integration

• Oil Holdup

4 3 2 1

3 2 1 0( )r x x a x a x a x a

2

0

( )

L

oV r x dx

2

oo

Vh

R L

L

x

Holdup correlation

7

0(1 ) (1 ) 0q m

os o ws o ref o oU h s U h cV h h

11

222 22 1

2 2

i

ii

i

n

nnn

ref i

gDDV a gD

7 3

4 2

i

i

nq

n

c,m, a, g, D, µ, ρ, n are constants From Rodriguez et al, 2006

Evaluation

Parameters

Fractional difference between areas (DFA)

Sensitivity (S)

Mean square error (MSE)

Number of sites of disagreement (NSD)

Holdup results

8

Holdup by

correlation

(Rodriguez et al,

2006)

Active Contours

Mathematical

Morphology

Holdup

Relative

error

% Holdup

Relative

error

%

Test 1

0.521 0.462 11.32 0.543 4.22

Test 2

0.812 0.625 23.03 0.804 0.98

Test 3

0.709 0.584 17.63 0.676 4.65

Test 4

0.814 0.693 14.86 0.797 2.09

Averag

e 16.71 2.98

Film thickness measurement using infrared images

• Rectangular channel with adjustable space

between the sensor and the wall in front of

it.

• Micrometer screws from 0 to 700 microns in

steps of 100 microns.

• Oil was added to the water and then

introduced as a flow on the top of the

channel

• The average measured values (using

conductance planar sensor) were compared

with the film thickness calculated by the

infrared images

9

Experiments

Test Vo (mL) Vw (mL) Average

Error [%]

Test 1 to Test 7

5 400

16.53

Test 8 to Test 14

10 400

20.68

Test 15 to Test 21

100 100

28.42

10

Im

Im

1( ) 100%S

N

F FARE abs

N F

Algorithm

11

Pre-processing:

Cutting

Segmentation:

Mathematical morphology

Film thickness

Calculation

Im

1WF N MP

NP

Vo

(mL)

Vw

(mL)

NP MP

5 400 11935 400

Results

12

10 20 30 40 50 60

5

10

15

0 100 200 300 400 500 600 700

Test 4. Micrometers position 400µm and average film

thickness by image 383.04µm and by sensor 373.28µm.

10 20 30 40 50 60

5

10

15

0 100 200 300 400 500 600 700

Test 14. Micrometers position 700µm and average film

thickness by image 613.7243µm and by sensor 611.295µm.

10 20 30 40 50 60

5

10

15

0 100 200 300 400 500 600 700

Test 20. Micrometers position 600µm and average film

thickness by image 531.53µm and by sensor

504.04µm.

Results

13

100 200 300 400 500 600 7000

100

200

300

400

500

600

700

Micrometers Position [m]

Wa

ter

Film

Th

ickn

ess

By Image Oil 5mL Water 400mL

By Sensor Oil 5mL Water 400mL

By Image Oil 10mL Water 400mL

By Sensor Oil 10mL Water 400mL

By Image Oil 100mL Water 100mL

By Sensor Oil 100mL Water 100mL

Planar sensor simulations

• Objetive: Find the best capacitive planar sensor geometry

• 5 geometries were tested

• Comparative features:

• maximum measurable thickness of the film,

• quasi-linear sensor characteristics

• spatial resolution.

• Calculation of the sensor characteristic is done by solving the

three-dimensional potential equation within the liquid film on

top of the sensor electrode system using Comsol.

14

Planar sensor geometries Geometry 1 Geometry 2 Geometry 3 Geometry 4

Geometry 5

15

Simulation results

100 200 300 400 500 600 700 800 900 1000

0.4

0.5

0.6

0.7

0.8

0.9

1

Film Thickness [m]

Dim

ensio

nle

ss C

apacitance [

-]

Geometry 1

Geometry 2

Geometry 3

Geometry 4

Geometry 5

16

NEXT STEPS

17

• Calibration of capacitive planar sensor

NEXT STEPS

18

•Sensor characterization: System Frequency response

•Sensor and camera setting-up in vertical 2” pipe

•Oil-Water tests (dispersed flow) in vertical pipe

•Water film thickness estimation by :

• Capacitive sensor

• High speed images

Acknowledgments

19