A SEGMENTED CAPACITANCE TOMOGRAPHY FOR VISUALISING

MATERIAL DISTRIBUTIONS IN PIPELINE CONVEYING CRUDE PALM OIL

ELMY JOHANA BINTI MOHAMAD

A thesis submitted in fulfilment of the

requirements for the award of the degree of

Doctor of Philosophy (Electrical Engineering)

Faculty of Electrical Engineering

Universiti Teknologi Malaysia

OCTOBER 2012

v

ABSTRACT

A segmented electrical capacitance tomography (ECT) imager for palm oil

process monitoring system is constructed and presented in this work.The goal of this

study is to use the process monitoring system as an instrument to upkeep the local

and foreign palm oil mill. This is to ensure that the monitoring of crude palm oil

(CPO) in conveying pipeline during extraction of palm oil mill process flow process

is efficiently controlled. The system has the capability to visualize the percentage of

liquid that exist within the vessel therefore the data can be utilized to design and

create better process equipment in mill process. It will also be used to control some

processes in order to boost the quality of crude palm oil and the POME (Palm Oil

Mill Effluent) treatment process. Most ECT in earlier research were created rapidly

and utilized well in multiphase flow measurement in numerous applications such as

in oil and gas industries, gas/solids cyclone, milk flows and fluidized beds.

Experimentally, this work investigates the capability of using a twin-plane

segmented ECT sensor with 16 portable electrodes using two differential excitation

potentials transmitted signal in order to recognize the concentration and velocity

profile as well as the phase concentration of crude palm oil related multiphase

systems (liquid and gas). The attained concentration profile which is received from

the capacitance measurements is capable to provide image of the liquid and gas

mixture in the pipeline therefore, the separation process (between oil and liquid

waste) becomes much easier and the crude palm oil‟s quality can be dependably

monitored. The visualization results deliver information regarding the flow regime,

superficial velocity and concentration distribution in two-phase flow-rate

measurement system incorporating a liquid flow measuring device. The information

obtained is able to help in the process equipment designing, verification of existing

computational modeling and simulation techniques. It may also assist in process

control and monitoring during the palm oil extraction process.

vi

ABSTRAK

Pengimej kapasitan elektrik bersegmen tomografi (ECT) bagi sistem paparan

proses kelapa sawit telah dibina dan dipersembahkan dalam kerja ini. Matlamat

sistem ini adalah untuk digunakan sebagai instrumen untuk mengekalkan kualiti

minyak sawit tempatan dan asing. Ini adalah untuk memastikan bahawa pemantauan

penghantaran aliran proses minyak sawit mentah menerusi saliran ketika proses

pengekstrakan minyak sawit dapat dikawal dengan lebih efektif. Sistem paparan ini

mempunyai keupayaan untuk memaparkan peratusan cecair yang wujud dalam

saliran, dengan itu data tersebut boleh digunakan untuk mereka bentuk dan mencipta

peralatan untuk proses yang lebih baik bagi kilang pemprosesan. Ia juga boleh

digunakan untuk mengawal beberapa proses untuk meningkatkan kualiti minyak

sawit mentah dan proses rawatan POME (Palm Oil Mill Sdn Efluen). Kebanyakan

ECT dalam penyelidikan awal telah dicipta dengan pantas dan digunakan dengan

baik bagi pengukuran aliran berbilang fasa dalam pelbagai aplikasi seperti industri

minyak dan gas, gas / pepejal siklon, aliran susu dan pepejal terbendalir. Kajian

penyelidikan ini akan menganalisa keupayaan pengesan ECT satah-berkembar

dengan 16 elektrod mudah alih dengan menggunakan dua isyarat pengujaan beza

upaya yang berlainan untuk mengenalpasti konsentrasi dan profil halaju serta

konsentrasi fasa minyak sawit mentah berbilang fasa dengan yang bersekutu

dengannya (cecair dan gas). Profil konsentrasi yang dikenal pasti daripada

pengukuran menerusi sistem tersebut mampu untuk memaparkan campuran gas

dalam saluran paip. Oleh itu proses pengasingan (minyak berasingan dan sisa cecair)

menjadi lebih mudah dan kualiti minyak sawit mentah boleh dipantau. Keputusan

visualisasi memaparkan maklumat mengenai aliran, halaju permukaan dan taburan

konsentrasi menerusi sistem pengukuran kadar aliran dua fasa yang digabungkan

dengan peranti pengukur aliran cecair. Maklumat ini dapat membantu dalam proses

mereka bentuk peralatan, pengesahan pemodelan pengiraan sedia ada dan teknik

simulasi. Ia juga dapat membantu dalam kawalan proses serta pemantauan sepanjang

proses perahan minyak sawit.

vii

TABLE OF CONTENTS

CHAPTER TITLE PAGE

DECLARATION ii

DEDICATION iii

ACKNOWLEDGEMENTS iv

ABSTRACT v

ABSTRAK vi

TABLE OF CONTENTS vii

LIST OF TABLES xiii

LIST OF FIGURES xv

LIST OF ABBREVIATIONS xxii

LIST OF SYMBOLS xxiv

LIST OF APPENDICES xxvi

1 INTRODUCTION 1

1.1 Introduction of Research Study 1

1.2 External or Internal ECT Sensors 6

1.3 Number of electrodes 8

1.4 Sixteen Segmented ECT Sensor Electrodes 9

1.5 A Twin Plane ECT 11

1.6 Cross-correlation velocity measurement 11

1.7 Background of Research Problem 14

1.8 Problem Statements 17

1.9 Research Objectives 18

1.10 Research Scope 19

1.11 Significant research contributions 20

viii

2 LITERATURE REVIEW 22

2.1 Research Background 22

2.2 Requirement Placed on Industrial Tomography 23

2.3 Classification of Tomography Imaging Principles 25

2.3.1 Hard Field Sensors 26

2.3.2 Soft Field Sensors 26

2.3.3 Hard Field vs. Soft Field Sensors 27

2.4 Sensors Principles for Industrial Tomography Imaging 27

2.5 Types of Images Reconstruction Algorithm 29

2.6 Linear Back Projection Reconstruction Algorithm (LBP) 30

2.7 Types of Projections 30

2.8 Related Work on ECT 32

2.9 System Design and Application 35

2.10 Summary 40

3 SEGMENTED ECT SENSOR MODELING AND 41

SIMULATION

3.1 Introduction 41

3.2 Forward Modeling Development 43

3.2.1 Forward Modeling Process Using COMSOL

Multiphysics Software 46

3.2.2 Simulation design process 47

3.2.3 Preparing for simulations 49

3.2.4 FEM Meshing 50

3.2.5 Set electrical properties domain and boundary

condition 50

3.2.6 Preparing of Excitation Electrode 52

3.2.7 Numerical calculation of the electrical field 54

3.3 Simulation of Segmented ECT model for Single Excitation

Potentials Schemes 56

3.3.1 Single excitation potentials, single-electrode

excitation for 16 sensor electrodes 57

3.3.2 Permittivity Distribution 59

3.3.3 Different Mixture of Permittivity Distribution 59

ix

3.3.4 The effect of increasing the size of permittivity using

single excitation potential,single-electrode excitation

schemes 61

3.4 Simulation of Segmented ECT model for Two Differential

Excitation Potentials Schemes 63

3.4.1 Two differential excitation potentials, single-electrode

excitation for 16 sensor electrodes 63

3.4.2 The analysis of increasing the diameter of the higher

dielectric material using two different excitation

potentials 65

3.5 Simulative Study on Image Reconstruction Algorithm 66

3.5.1 Sensitivity Distribution 68

3.5.2 Sensitivity Map 69

3.5.3 Forward problem and linearization solution 77

3.5.3.1 Normalization of capacitance measurements 77

3.5.3.2 Normalized sensitivity and permittivity 80

3.5.3.3 Process normalizing 81

3.5.4 Concentration profile 82

3.5.5 Linear Back Projection (LBP) Algorithm 84

3.6 ECT Image Reconstruction Simulator 86

3.7 Simulative Velocity Measurement for Twin plane Segmented

ECT Sensor 93

3.7.1 Cross-correlation Principle 93

3.7.2 Simulative Velocity Measurement using Cross

Correlation Technique 96

3.7.3 Mean Correlation 96

3.7.4 2-D Correlation Coefficient 99

3.8 Summary 100

4 SEGMENTED ECT HARDWARE AND SOFTWARE 106

DEVELOPMENT

4.1 Introduction 106

4.2 ECT Segmented Hardware Process Development 107

4.3 Basic Principle of ECT system 111

x

4.4 ECT Sensor Design Configuration 114

4.5 Electrode Sensor Design 116

4.5.1 Length and Diameter of Sensor Electrodes 116

4.5.2 Driven Guard Electrodes 118

4.5.3 Earthed Screens 120

4.6 Electrode Connecting Techniques 121

4.7 Gripper and Handle Model 123

4.8 Pipelines 124

4.9 Sensor Measurement Circuit 125

4.9.1 Switching Circuit 128

4.9.2 Capacitance Measurement Circuit 129

4.9.3 AC-to-DC Converter Circuit 131

4.9.3.1 Absolute Value Circuit 132

4.9.3.2 Low-Pass Filter 133

4.9.4 Programmable Gain Amplifier (PGA) 134

4.9.5 Analog to Digital Converter (ADC) 135

4.9.6 Microcontroller PIC16F87 136

4.10 Main Controller Unit 138

4.10.1 Microcontroller PIC18F4550 138

4.10.2 Function Generator 140

4.10.2.1 Function Generator 142

4.10.3 Interfacing between Sensor to Computer 144

4.10.3.1 Data Retrieving in PC 146

4.11 ECT Segmented Software Process Development 147

4.11.1 Sensing Modules Firmware 148

4.11.2 Main Control Unit Firmware 150

4.12 Graphical User Interface (GUI) 152

4.12.1 Bar Graph Analysis for Liquid Level Concentration 155

4.12.2 Bar Graph Cross Correlation for Superficial Liquid

Velocity 156

4.13 ECT Measurement Programming Modules 157

4.14 ECT System Concentration Measurement 157

4.15 Error Measurement in ECT System 158

4.16 Summary 159

xi

5 EXPERIMENTAL SETUP, RESULT AND ANALYSIS 164

5.1 Research Background 164

5.2 Experimental setup for standing capacitance

measurement 165

5.2.1 Comparisons of standing capacitance

measurements of different types of dielectric

materials (air, CPO, sludge, and water) 166

5.2.2 Standing capacitance measurements of different

phantom sizes (flow regime) 170

5.3 Experiments performed to compare the distribution of

potentials 171

5.3.1 Single excitation potential technique 171

5.3.2 Two differential excitation potential techniques 173

5.4 Signal-to-noise ratio 175

5.5 System calibration 178

5.6 Core sensitivity analysis 183

5.7 Two Phase Flow Visualization Analysis 190

5.8 Dispersions flows of oil -water 192

5.9 Dispersions flows of oil-sludge 196

5.10 Repeatability of Image Concentration Measurement for

ECT System 199

5.11 Velocity Profile Experimental Arrangement 201

5.11.1 Velocity profile observation for horizontal flow

measurement 205

5.11.2 Velocity profile observation for horizontal

downward inclined flow (declined ) 20˚ and 30˚ 209

5.11.3 Velocity profile observation for horizontal upward

inclined flow (inclined ) 20˚ and 30˚ 212

5.12 Velocity Profile Measurement Result Analysis 215

5.13 Summary 217

6 CONCLUSIONS AND SUGGESTIONS FOR FUTURE 220

WORK

6.1 Conclusions 220

xii

6.2 Significant Contributions towards Process Tomography 223

6.3 Recommendations for Future Work 224

REFERENCES 226

Appendices A – D 235-264

xiii

LIST OF TABLES

TABLE NO. TITLE PAGE

1.1 The Relationship between Total of Electrodes used and Total

of Independent Measurements from Existing ECT Systems 8

2.1 Sensor for industrial tomography imaging. 29

2.2 Existing ECT System in Industrial Application 33

3.1 Physical Sensor Parameter 48

4.1 Digital data transferred to the PC for image reconstruction 146

5.1 Standing capacitances of air, CPO, sludge, and water 168

5.2 Standing capacitance (pF) caused by the different sizes (mm)

of dielectric materials 171

5.3 Lowest limit voltage sensor reading when the sensor was

filled with gas during calibration 181

5.4 Highest limit voltage sensor reading when the sensor was filled

with water during calibration 181

5.5 Comparison of actual and measured readings for the

concentration measurement (C.K. Seong, 2008) 183

5.6 Actual and measured readings for the proposed 185

5.7 Actual concentration measurements 188

5.8 Reconstructed image of horizontal oil–water flow concentration 194

5.9 Reconstructed image of the horizontal oil-sludge flow

concentration 197

5.10 ANOVA and Gage R&R 199

5.11 Horizontal velocity profile measurement 206

5.12 Declined 20˚ velocity profile measurement 211

5.13 Declined 30˚velocity profile measurement 211

xiv

5.14 Inclined 20˚velocity profile measurement 213

5.15 Inclined 30˚velocity profile measurement 214

xv

LIST OF FIGURES

FIGURE NO. TITLE PAGE

1.0 A schematic diagram of an ECT system 5

1.1 Flow velocity measurement using cross-correlation 13

1.2 Biological ponding system of Palm Oil Milling, Sime Darby

Research Sdn. Bhd. Carey Island, Banting Selangor, Malaysia,

2011. 15

1.3 (a) Flow Diagram of Palm Oil Milling and (b) Crude Palm

Oil Process -Sime Darby Research Sdn Bhd, Carey Island,

Banting Selangor, Malaysia, 2011. 16

1.4 Clarifier Tank 16

2.1 Various form of projections through a cross-section of

conveyor, pipe-line or vessel 31

3.0 (a) Schematic representation of the measurement principle of

an ECT system (b) electric field lines that exist between any

two electrodes 45

3.1 Illustration of sixteen sectors of portable ECT 48

3.2 a) 2D segmented geometry of portable ECT sensor b) 3D

geometry of portable ECT sensor 49

3.3 FEM Meshing; (a) initial mesh (b) finer mesh 50

3.4 Allocate the boundary and subdomain settings in COMSOL 51

3.5 (a) Potential Distribution and (b) Electrical filed for single

excited electrode. The electrical potential and electrical field is

encoded with the color 53

3.6 Potential lines for single excited electrode. The electrical field

lines are calculated for electrode 1 56

xvi

3.7 16 segmented sensor arrangement 57

3.8 Simulation result shows the electric field lines are deflected

depending on material distribution, when one electrode was

excited 58

3.9 Results from (a) 2D and (b) 3D simulation, respectively. Blue

area is sensing region represent air (εair =1), and red area

represents water (εwater =80). 59

3.10 Permittivity distribution for different types of dielectric (a)-(f) 60

3.11 Image simulations when increasing the diameter of dielectric

materials using single potentials/voltage source. 61

3.12 Simulated capacitances due to increasing the size of permittivity

of the dielectric material, ℰr=80, using single potentials/

voltage source schemes 62

3.13 Electrical field distribution for 2D when half of electrode pairs

was excited with low voltage source and another half with

higher voltage source injected 64

3.14 Images simulation when increasing the diameter of dielectric

material using two different potentials/voltage sources. 65

3.15 Simulated capacitances due to increasing of the permittivity

size of the dielectric material, εr=80, using two different

potentials/voltage source schemes. 66

3.16 Back projecting an image 66 67

3.17 The sensitivity matrix of capacitance measured between

electrodes 69

3.18 32x32 square sensitivity matrix 70

3.19 Generation of sensitivity map in (a) 2D and (b) 3D. 71

3.20 Sensitivity maps for 16 projections (a)-(c) 72

3.20 Sensitivity maps for 16 projections (d)-(f) 73

3.20 Sensitivity maps for 16 projections (g)-(i) 74

3.20 Sensitivity maps for 16 projections (j)-(l) 75

3.20 Sensitivity maps for 16 projections (m)-(n) 76

3.21 ECT sensors with a high permittivity material in the air 79

3.22 Series capacitance model in ECT 79

3.23 Graphical user interfaces for measurement process simulation 87

xvii

3.24 Image reconstruction simulation with single excitation

potential for one liquid droplet in the center area (a) before

normalized and (b)after normalized 88

3.25 (a) Image reconstruction simulation for stratified flow

(b) comparison of image concentration percentage (%) with

basic LBP and after normalized 89

3.26 (a) Sensitivity distribution of single excitation potential and

(b) normalized sensitivity distribution 90

3.27 Simulation result of image reconstruction for different types

of flow regimes and permittivity distribution. 92

3.28 The principle of velocity measurement using cross correlation

technique 94

3.29 The movement of the flow through upstream sensing area 97

3.30 The movement of the flow across at downstream sensing area 98

3.31 Mean values plotted for each frame recorded at upstream

sensing area 98

3.32 Mean values plotted for each frame recorded at downstream

sensing area 98

3.33 Cross-correlation signals between upstream and downstream 99

3.34 Signal cross-correlation between upstream and downstream

using 2-D Correlation Coefficient 100

4.0 A twin-plane arrangement of segmented ECT sensor electrodes 107

4.1 Portable segmented electrode sensor arrangement mounted

symmetrically on the pipe wall 108

4.2 Overview of the system 109

4.3 Topology of the segmented ECT System 110

4.4 Capacitance measurement principle 111

4.5 Cross-sectional view of a typical ECT sensor with 12 electrodes 113

4.6 Cross-sectional view of the ECT sensor with 16-segmented

portable electrodes 115

4.7 Dimensions of the electrode: (a) previous design measured at

100mm and (b) new design measured at 120mm in length of

the sensor area 118

xviii

4.8 (a) ECT sensors with end guard and axial guard, (b) ECT

sensors with driven guard 119

4.9 The earthed screen is placed at the top layer of the electrode 121

4.10 Electrode plates with mounted PCB sockets 122

4.11 Sensing module 122

4.12 Customized design of the gripper model 123

4.13 Hardware design of the sensor jig of the ECT system 124

4.14 Block diagram of the signal conditioning module 126

4.15 Signal conditioning circuit module 126

4.16 Complete portable electrode sensor module 127

4.17 Portable electrodes interconnected by a 26-way IDC cable 127

4.18 The circuit arrangement of four switches for electrode

selection 128

4.19 Detector and AC amplifier circuits 130

4.20 The absolute value circuit 132

4.21 Application of the low-pass filter on the absolute value circuit 133

4.22 Programmable gain instrumentation amplifier (PGA) 135

4.23 ADC10461 provided by National Semiconductor 136

4.24 Microcontroller PIC16F876 circuit diagram 137

4.25 Microcontroller PIC18F4550 circuit diagram 139

4.26 The main control unit for twin plane segmented ECT 139

4.27 Connection of XR2206 141

4.28 Op-Amp TL081 to produce two different waveforms 24Vp-p

and 4Vp-p 142

4.29 Transmitted signal from the signal generator with single

excitation potentials/voltage source 143

4.30 Sequentially transmitted signal from the signal generator with

two difference potentials/voltage sources 144

4.31 Output waveform from the Function Generator 144

4.32 Features in a PIC18F4550 147

4.33 Flowchart programming for the sensing modules 149

4.35 ECT system main GUI 152

4.36 Liquid level and cross correlation online monitoring system 153

xix

4.37 Bar graph for displaying liquid level concentration and

superficial liquid velocity 154

4.38 Basic flow chart for the GUI ECT system 155

4.39 Cross correlation for liquid velocity 156

5.0 Digital capacitance meter (MODEL 3000) 165

5.1 Inter-electrode capacitance reading using the digital capacitance

meter monitoring system of GLK INSTRUMENTS 166

5.2 Pipeline was filled with (a) water (b) sludge and (c) CPO 167

5.3 Comparisons of the standing capacitances of air, CPO, sludge,

and water 169

5.4 Portable segmented ECT with bottle and test tube used to create

phantoms inside the ECT sensor. The bottle and test tube were

filled with water, creating a phantom of annular water flow 170

5.5 Transmitted signal from the signal generator using a single

excitation potential/voltage source 172

5.6 Potential distribution (Vpp) for the low permittivity (phantom

was empty) and high permittivity (phantom was filled with

water) in the pipeline using a single potential excitation 172

5.7 Transmitted signal from the signal generator with two different

potential /voltage sources sequential 174

5.8 Potential distribution (binary number) after normalization for

the low permittivity (phantom was empty) and high permittivity

(phantom was filled with water) in the pipeline using two

differential excitation potentials. 174

5.9 New ECT design approach with a 120mm long sensor area

(a) voltage output reading for low permittivity (phantom was

empty) and (b) for high permittivity (phantom was filled with

water) in the pipeline using two differential excitation

potentials 176

5.10 Previous ECT design with a 100 mm long sensor area

(a) voltage output reading for low permittivity (phantom was

empty) and (b) for high permittivity (phantom was filled with

water) in the pipeline using a single excitation potential 177

xx

5.11 Comparisons of the reconstructed images and liquid levels

using the two differential and single excitation techniques 178

5.12 A test section for calibrating the sensor prior to measurements 180

5.13 Reconstructed image for both planes (Planes 1 and 2) during

calibration when the sensor was filled with (a) gas (low-

permittivity material) and (b) water (high-permittivity material) 182

5.14 Core measurements (a)–(f) (C.K. Seong, 2008) 184

5.15 Reconstructed images from the containers of different sizes 187

5.16 Comparisons of the actual and measured data collected for

(a) water and (b) gas concentrations. 189

5.17 Phase distribution in a pipe cross-section for mixture with

input oil fraction 70% at different angles of inclination taken

from N.M. Hasan et al. (2007) 191

5.18 Horizontal oil-water/sludge flow rigs 192

5.19 Horizontal oil–water flows 192

5.20 Horizontal oil–sludge flows 196

5.21 (a)-(d) Gage R&R For ECT Image Concentration Measurement 200

5.22 Liquid-gas flow rig system 203

5.23 A twin-plane segmented ECT mounted onto the acrylic pipe

of the test section 203

5.24 Distance between the sensing planes 204

5.25 Axial View of sensor structure taken from Hayes, D.G., et al.,

(1995) 205

5.26 Measurement using liquid-gas flow rig system in horizontal 206

5.27 Velocity profile observation for horizontal flow measurement 207

5.28 Reconstructed image tomogram for liquid volumes of (a) 15

LPM and (b) 25 LPM 208

5.29 The measurement using liquid-gas flow rig system in

downward inclined flow (a) 20˚ and (b) 30˚ 210

5.30 Velocity profile observation for horizontal downward inclined

flow (declined) in 20˚and 30˚ 211

5.31 The measurement using liquid-gas flow rig system in upward

inclined flow (a) 20˚ and (b) 30˚ 213

xxi

5.32 Velocity profile observation for horizontal upward inclined

flow(inclined) in 20˚and 30˚ 214

5.33 Capability analysis data for velocity measurement 216

xxii

LIST OF ABBREVIATIONS

- degree

AC - Alternative-Current

ADC - analog to digital converter

ANOVA - Analysis of variance

CMOS - complementary metal oxide semiconductor

CpK - Process capability

DAC - digital to analog converter

DAS - Data acquisition system

ECT - Electrical Capacitance Tomography

EEPROM - Electrical Erasable Programmable Read Only Memory

EIT - Electrical Impedance Tomography

ESD - electrostatic discharge

fps - frame per second

GUI - Graphical user interface

GR&R - Gauge of repeatability and reproducibility

Hz - Hertz

I/O - Input / Output

IDC - insulation displacement contact

IOCTL - input and output control

IPT - Industrial Process Tomography

ISA - Industry Standard Architecture

kΩ - kilo-ohm

kbps - kilobit per second

kHz - kilohertz

LBP - Linear Back Projection

xxiii

LED - light emitting diode

MATLAB - Matrix Laboratory

Max. - Maximum

Mbps - Megabit per second

MHz - MegaHertz

mm - millimetres

MOR - Model Based Reconstruction

MRI - magnetic resonance imaging

ms - millisecond

MSIRT - Multiplicative simultaneous iterative reconstruction technique

NMR - Nuclear Magnetic Resonance

NMRT - Nuclear Magnetic Resonance Tomography

OIOR - Offline iteration and online reconstruction

op-amp - operational amplifier

PC - Personal Computer

PCB - Printed Circuit Board

PET - Positron Emission Tomography

PSNR - peak signal-to-noise ratio

PT - Process Tomography

PTL - Process Tomography Limited

RAM - Random Access Memory

RMSE - root mean square error

SNR - Signal to Noise Ratio

SPI - Serial Peripheral Interface

SW - Switch

USB - Universal Serial Bus

Vp-p - Voltage peak-to-peak

xxiv

LIST OF SYMBOLS

A - Total gain of measurement system

A/D - Analog to digital

C - Matrix of inter-electrode capacitance

CH - Capacitance measured at higher permittivity

CL - Capacitance measured at lower permittivity

CM - Measured capacitance

CN - Normalized capacitance

Coil - Relative capacitance of oil

Cr - Relative capacitance

Cs1 - Stray capacitance of connecting lead

Cs2 - Stray capacitance at Op-Amp feedback point

Cwater - Relative capacitance of water

Cx - unknown standing capacitance

d - Distance of 2 parallel plate

D - Sensor diameter

ε - Effective permittivity

ε0 - Permittivity of free space

εo - Relative permittivity of oil

εr - Relative permittivity

εw - Relative permittivity of water

f - Frequency

funitygain - Unity gain frequency

K - Matrix of permittivity

Ke - Effective pixel permittivity

Ken - Normalized effective pixel permittivity

xxv

KH - Pixel permittivity at lower permittivity

KL - Pixel permittivity at higher permittivity

L - Length of electrode

m - Number of individual standing capacitance

M - Total number of pixels

n - Total number of pixels

N - Number of measuring electrodes

Q - Unknown matrix

S - Sensitivity matrix

S-1

- Inverse sensitivity matrix

SNR - Signal to Noise Ratio

ST - Transpose sensitivity matrix

Vi - Input voltage

Vo - Output voltage

VR - Volume ratio

W - Width of electrode

x - Volume ratio

ΔC - Error capacitance matrix

ΔK - Error pixel matrix

ωo - Corner frequency

xxvi

LIST OF APPENDICES

APPENDIX TITLE PAGE

A Paper Published/Presented & Awards 235

B Sensitivity Matrix of Capacitance 242

C(I) Sensor Jig Design 247

C(II) Sensor Measurement Schematic Diagram 248

C(III) Main Controller Schematic Diagram 249

D(I) Standing Capacitance Reading of Crude Palm Oil 250

D(II) Liquid Gas Flow Rig System Design 253

D(III) Tomogram of Liquid Velocity Profile 255

CHAPTER 1

INTRODUCTION

1.1 Introduction of Research Study

It has been awhile since the introduction of Process Tomography. Ever since

the very first time of its establishment in 1950‟s, the development of tomographic

instrumentation has led to the widespread of body scanners obtainability which has

become part of modern medicine requirements. Process Tomography has evolved

essentially during the mid-1980‟s. A number of imaging equipment for processes

was termed in the 1970‟s, but this involved using ionization in x-rays and many more

in general. Meanwhile, in the mid-1980‟s, the present invention was launched in

which it has resulted in the recent work on process tomography systems.

In the very beginning, tomography was practiced in medical field for

diagnostic purposes. In medical applications, the radiation source, X-rays are utilized

to form images of bones based on their attenuation coefficient. This approach lets us

observe the internal structure (bones) of our body without requiring us to dissect or

cut it open. Thus, tomography approach is also identified as a non-invasive method

to examine the internal structure or behavior of a material. It has to be highlighted,

however, that the tomography concept and its non-invasive way of imaging are not

constrained to the medical field alone. This fundamental difference results

2

in differences in sensor design, imaging speed, image reconstruction algorithms and

also cost.

The tomographs required from „softfield‟ tomographic techniques are found

to be more semi-quantitative or qualitative and have a low spatial resolution where

on the other hand, a „hardfield‟ technique results in more quantitative measurements

with higher spatial resolution. Nevertheless, conventional x-ray tomography

generally has lower temporal resolution unlike ECT. Thus, ECT is helpful in the

evaluation of rapid changes in hydrodynamic (Chaplin, G. et al, 2005).

In simple terms, tomography is an imaging technique which allows one to

define a closed system contents without physically needing the individual to look

inside it. It is also a technique utilized to examine the internal behaviors of flowing

materials inside a pipeline. Process Tomography can be identified as imaging

process parameters in space and time. Important flow information such as

concentration measurement, velocity, flow rate, flow compositions and others can be

acquired without having to interrupt the process or object. It is also used to delineate

the internal composition of pipes or mixing vessels. The sensor signals are filtered,

amplified, digitized and processed in a computer as a final stage with the use of

certain image reconstruction algorithm in order to construct cross-sectional image.

Usually, many of the flow information are based on the cross-sectional image

detected. For example, the cross-sectional image itself already makes available the

flow material‟s concentration.

Principally, Process Tomography can be separated into two categories which

are electrical tomography and optical tomography. There are few types of electrical

tomography, EIT (electrical impedance tomography), ECT (electrical capacitance

tomography) and EMT (electromagnetic tomography). Based on electromagnetic

field theory, Electrical Tomography delivers inexpensive non-intrusive imaging

systems with low but sufficient resolution of the internal distributions of processes.

Optical fiber transmitters and receivers are utilized by optical tomography system,

3

such as fan beam optical tomography system and parallel beam optical tomography

system.

Process Tomography compromises a convenient means of model verification

in an industrial environment, unlike in a simplified 'model' reactor with the use of

conventional laser or optical tracer techniques. Conventional techniques need

invasive sampling methods or that the process mixture has to be modified in some

artificial manner (eg solid/liquid suspensions need to be diluted so that optical access

for measurements can be obtained). Moreover, the measurement information is

usually limited to one small zone within the process vessel, or entails a multiplicity

of measurement zones, while tomographic techniques have an exceptional spatial

range and can be utilized for 2- and 3-dimensions imaging.

In addition to model validation, the capability to figure the profile‟s

component concentration and in some cases to recognize phase sizes and boundaries

within vessels and pipelines will somehow offers a series of information. This

information is valuable for Process Engineers in shedding some light on fundamental

reaction kinetics as well as for optimum geometric design of large scale equipment.

As 3-dimensional imaging techniques build up, it is predicted that the process

tomography will be employed for the reasons of plant controlling which can either be

for alarm functions or in the form of a full mass balancing and circuit monitoring

service.

An extensive range of applications for process tomography is currently being

developed actively. Electrical engineers seek to re-design the instruments in a

fundamental way for special tasks in some applications and some of the examples are

like the combustion images at up to 36,000 frames per second, environmental

groundwater imaging, separation process and hydrocyclone imaging, stirred reactors

imaging, fault detection in rotating components, multiphase flow measurement and

control from tomographic measurements.

4

There is a huge request for direct analysis of the internal characteristics of

process plants with regards to enhancing the equipment operation and design.

Process tomography may as well assist in the equipment design to keep control the

optimum flow conditions. In other word, process tomography can be categorized into

five main categories as follows:

i. Characterization of individual components and non-destructive testing.

ii. Modeling in a laboratory environment.

iii. Equipment design and optimization in a laboratory or industrial

environment.

iv. Process monitoring and control in an industrial environment.

v. Remote sensing in manufacturing, quality control, environment protection

or pollution control on an industrial or waste containment.

The industrial sector has seen the booming popularity of Electrical

Capacitance Tomography (ECT) systems each day. In general, this technique can be

defined as the analysing process which takes place within the pipes and chemical

vessels through the examination of the internal distribution of permittivity. One of

the clear examples of the use of ECT is to detect the flow regime and the degree of

entrainment. Unlike the conventional methods, the flow rate of the solids can be

determined more precisely with ECT thus enhancing the control and the

manufacturing operation‟s efficiency. It has been utilized in various applications

especially in chemical and petroleum industry, including multiphase flow in oil

pipeline.

ECT has been employed in this research to monitor the concentration of

crude palm oil. The system is intended to be utilized as an instrument to ensure both

local as well as foreign palm oil mills so that the control in monitoring the crude

palm oil‟s (CPO) quality flow in conveying pipelines during extraction of palm oil

mill process cefficiently done. The visualization results impart information regarding

the flow regime, superficial velocity and concentration distribution in two-phase

flow-rate measurement system incorporating with liquid flow measuring device.

5

The ECT system consists of a sensor, control computer as well as capacitance

measuring unit. The sensor contains a collection of electrodes wound which are to be

imaged located all over the periphery of the pipe or the vessel. The control computer

will receive the measuring unit conditions the signal acquired from the sensor. The

information received will be processed by the control computer processor which will

also undergo the creating of permittivity distribution images matching to the cross-

section as seen by the sensors. To minimize the measurement errors, all parameters

which were measured and calculated are normalized. The Linear Back Projection

Algorithm (LBPA) is applied by the control computer to generate permittivity

distribution images. The images will then be shown on a 1024 square pixel grid with

a proper graduated colour scale to indicate the permittivity variation (Donthi, S.S.,

2004).

The images are analysed to figure the parameters associated with the process

such as spatial distribution and volume ratios of the materials within the pipe as well

as the flow velocities. ECT is applied in multi-phase flow-meters to inspect the

multiple fluids flow regime flowing in a pipe. Proper control signals are produced by

the permittivity images which determine the future course of a chemical process. The

ECT system shown in Figure 1.0 is used to get the permittivity distribution images of

contents of the pipe during a chemical process.

sensor

object

Figure 1.0 A schematic diagram of an ECT system

Sensing Module (Signal

Conditioning

Circuit)

Measured data

Control signal

Real time image reconstruction

using a computer

Computer

6

Figure 1.0 depicts a schematic diagram of an ECT system with its main

components. The system comprises of the following:

(i) Capacitance sensor which contains a collection of electrodes affixed to

periphery of the pipe which is to be imaged.

(ii) Capacitance measuring unit to obtain and process the signals receive

from the capacitance sensor.

(iii) Control computer to rebuild and show the permittivity distribution image

with the data received, and to observe and control the process happening

within the pipe.

1.2 External or Internal ECT Sensors

The capacitance electrodes can be mounted either inside or outside the vessel.

The electrodes will usually be mounted on the outside surface of the pipe or vessel if

the wall of the vessel is an electrical insulator and example of an electrical insulator

is plastic. In this case, the ECT sensor is believed to be non-invasive and non-

intrusive when the sensing electrodes are not in physical contact with the medium

within. Trevor York (2001) mentioned that the ECT sensor is non-intrusive if the

sensing electrodes are mounted within the pipe and are in contact with the medium

but not disturbing the flow at all means. Nonetheless, it is reported that most ECT

technique is a practical non-invasive technique for the multiphase flows

measurement (F. Wang, et. al., 2009).

In the case of a sensor with internal electrodes, the capacitance components

owing to the electric field within the sensor will constantly rise correspondingly to

the material permittivity when the sensor is occupied equally with higher permittivity

material. The wall gives a negative effect on the internal capacitance measurements.

This is all due to the reason that the wall capacitance is well in sequences with the

internal capacitance. However, the permittivity of the wall will trigger non-linear

7

changes in capacitance for sensors with external electrodes, which may increase or

decrease based on the thickness of the wall and the permittivity of the sensor wall

and contents (Daoye, Y. et al,, 2009).

There are only few techniques proposed in earlier studies which focus to

improve the uniform sensitivity distribution in the central area as well as to make

progress of the detection signal levels in the ECT system by using single electrode

excitation and one voltage source. Hence, this work aims to study the sensitivity

distribution and capacitance non-linear changes. A two different potential

excitation/voltage source is presented instead of using only one potential

excitation/voltage source to an excitation electrode at a specific time allocated. The

two different potential excitation/voltage sources is employed in sequence to

different excited electrode pairs to create excitation field which is considerably

uniform across the sensor. The image reconstruction simulations are given to reveal

these techniques capability of improving the sensitivity distribution and potential

distribution in the central area. Formerly, these techniques have never been applied

in an ECT system. The SNR (signal noise to ratio) can be improved proportionally

with the increase in voltage across the centre of the pipe unlike the one attained using

standard single excitation potential schemes.

In order to prove this technique, the advancing model using COMSOL

Multhiphysics is built to simulate the changes in capacitance between opposing

electrodes and the dielectric material permittivity. D The increasing diameter of the

higher permittivity insert using single and two different excitation potentials was

further discussed in Chapter 3.

8

1.3 Number of electrodes

The important aspect which has to be determined first when designing an

ECT sensor is the number of electrodes. An ECT is comprised of a number of

electrodes. Reported ECT system is mostly comprised of 8, 12 or 16 electrodes. The

size of each electrode is seen to decrease relatively to the growing number of ECT

sensor electrode from 8, 12, 16 or larger.

A small number of electrodes will give benefits of (1) A smaller number of

data acquisition channels is needed which considerably make simpler for the

hardware if every channels is run in parallel, (2) A faster data acquisition rate is

anticipated due to the reduction of capacitance measurements number. However,

with just a small number of electrodes, it is hard to expect a good image because the

number of independent capacitance measurements is small. On the other hand, if the

number of electrodes increases, an image with improved resolution can be expected

because the total of independent measurements will increase. Yet again, W.Q. Yang,

(2010) mentioned that some difficulties might occur if the number of electrodes is

too large such as (1) complicated and expensive hardware, (2) smaller capacitance to

be measured and (3) more measurements taken which result in a slower data

acquisition rate. Table 1.1 presents some matching figures for the number of

electrodes being 6, 8, 12 and 16.

Table 1.1: The Relationship between Total of Electrodes used and Total of

Independent Measurements from Existing ECT Systems

Total of electrodes used Total independent measurement Application

6 15 Combustion flame in an

engine cylinder

8 28 Wet gas separator

12 66 Gas-oil-water

16 120 Nylon polymerization

9

Measurement sensitivity of capacitance sensor is proportional to the electrode

area. Therefore, the increase in the number of electrode is mainly to enhance SNR

(Signal Noise to Ratio). The size of electrode has to be increased in order to improve

SNR because when the electrode size reduces, the SNR will also be reduced.

Therefore, image with better resolution can be obtained if the number of electrode is

increased by reducing the electrode size. (C.K. Seong., 2008).

Nevertheless, high resolution images can be obtained if the number of

electrode is high. High number of electrodes means the size of the electrodes is

smaller. If the size is small, measurement sensitivity will be lower unlike sensor with

few electrodes. Bigger electrodes promote higher sensitivity in the sensor but it will

reduce the resolution quality. If high axial resolution is needed, a small number of

short electrodes can be used together with separately excited axial guard electrodes.

This can stop the electric field from dispersing overly at each end of the sensor

electrodes. Most applications developed are using sensors with 8, 12, or 16

electrodes up till today. (Flores, N. et al., 2005). Therefore, a 16 segmented sensor

electrodes was chosen to be mounted symmetrically on the outer surface of an

insulating vertical pipeline for this particular work.

1.4 Sixteen Segmented ECT Sensor Electrodes

The Electrical Capacitance Tomography System of earlier inventions is set to

be fixed on a vessel. Functioning as a sensor, the electrode plates are permanently

assembled on the pipeline as this is typically distinctive for each new application.

When the installation is done, it will be impossible for the sensor to be removed and

thus, the distance between the sensing electronics and these sensor plates will usually

be made far

10

This portable system aims to permit the system smoothness to be assembled

and relocated from a pipeline to another. The system is specifically designed to

accommodate pipeline of various diameter sizes as well as flexibility of electrodes

sensor usage numbers relying on the pipeline size without having to redesign the

electrodes sensor. This sensor can function independently. Apart from that, the

driven guard that normally put between adjacent measuring electrodes and earth

screen which is mounted at the third layer beforehand has been fixed on the

segmented electrode sensor plates. The signal conditioning board is able to be

stretched in accordance with the pipe diameter because the cable noise and the

electrode are eliminated.

The sensing module comprises of integrated intelligent electrode sensing

circuits on each electrode sensors. A microcontroller unit and Data Acquisition

(DAQ) system is embedded on the electrode sensing circuit and the data acquisition

system is applied with the Universal Serial Bus technology (USB) technology that

allows the sensor to function self-sufficiently. The USB is also purposely applied to

experience high data transfer rate in data transferring. In order for the USB data

transfer rate to perform at full-speed, a microcontroller which functions as the

centralization control unit is used.

In this research, a longer sensor electrode has been expand to 120mm in

comparison to the earlier design by C.K. Seong, (2008) where the sensor electrode is

expanded only to 100mm where the capacitance measurement is only achieved at

femtofarad. The capacitance values are normally in a range of 0.5 and 0.01 pF as

according to W.Q. Yang, (2010). As a result to this, the reading of the length of

electrode is required to be increased in order to increase the capacitance. In this

research, the capacitance reading is well increased to picoFarad. The values of the

capacitance were confirmed in different types of dielectrics which provides reading

in a range of 0.1pF to 0.45 pF.

11

1.5 A Twin Plane ECT

The measurement techniques to get the volumetric flow rate for single –phase

flows are properly built with accuracy of lower than 1%. However, the multiphase

systems require some advanced technology concerning the concentration of one

phase in another. It was informed that the application of twin-plane tomographic

systems has potential to offer substantial data in multiphase flows. It promotes a

technique which is considerably attractive to measure the concentration and velocity

distributions, such as a flow meter. Hayes, D.G (1994) mentioned that the

combination of tomography and cross correlation techniques offers an opportunity to

measure the velocity profile. A twin plane ECT sensor was applied for velocity

profile measurement (W.Q. Yang, 1995, 1998, 2004, W. Warsito and L.S. Fan 2001,

S. Liu, et. al, 2002, 2005 and. H.G. Wang, et. al, 2007)

This research aims to create a twin plane of 16-segmented sensor electrode

where two sets of segmented sensor electrodes are mounted with one as up-stream

sensor and the other as a down-stream sensor. The design of electrode sensor has

seen some improvements and there is a proposal of a new approach to switch the

excitation potentials as transmitted signal to the electrodes. The research

development are to experimentally study the aptitude of using a twin-plane 16-

segmentd ECT sensor with 16 portable electrodes which uses two differential

excitation potentials transmitted signal to find the flow pattern, the phase

concentration and the velocity profile of crude palm oil related multiphase systems

(crude palm oil, water, sludge and gas).

1.6 Cross-correlation velocity measurement

Many of researchers have applied cross correlation techniques with a single

pair of transducers or twin-plane sensors for flow velocity profile measurement

12

(W.Q. Yang, 2000, 2010). In order to get velocity profile, the void or concentration

information between the two images in planes 1 and 2 has to be cross-correlated as

shown in Figure 1.2 (Dyakowski T. et. al, 2000).

The cross-correlation flow velocity measurement‟s important principles are as

stated in Figure 1.1 (a), is the signals „tagging‟ created by the fluid turbulence or

suspended particles moving inside the pipe. The sensors, which detect these signals

in empathy with the turbulence of the fluid or suspended particles, could be based on

many techniques such as the usage of ultrasound, electrical conductivity, optical

beams and electrical capacitance. In an ideal case, if a signal detected by the

upstream sensor reappears at the downstream sensor after a certain period τm (see

Figure 1.1 (b) ) and the distance ds between the two sensors is identified, the velocity

V can be determined as;

V = ds/ τm …..(1.1)

Hypothetically, the patterns or signals created by the fluid turbulence or

suspended particles will slowly shift on moving downstream. However, if the

downstream sensor position is practically near to the upstream sensor, the patterns or

signals would be sufficiently similar to each other, permitting the measurement of the

transit time τm (S. Liu, 2005).

It is essential to select the right distance between the two planes for cross-

correlation. Better similarity between the signals from the two planes can be gained

just by a smaller distance. Nevertheless, longer distance is needed by the

comparatively slow data acquisition and image reconstruction rates and the

interference between the two planes. Thus, it is important to take into account both

signals similarity as well as the dynamic behaviour of the system. A cross-correlation

function for the flow rate measurement can be calculated from two sets of signals;

13

…..(1.2)

where x(i ) and y(i ) are the up-stream and down-stream signals, N is the number of

samples in the summation, M is the number of samples in the cross-correlation

calculation and j is the number of the delayed sample (i.e. time delay). The transit

time m can be acquired by seeking out the peak of the cross-correlation function.

tjm . …..(1.3)

where j matches the peak of the cross-correlation function and t is the time interval

between each two samplings. Then, the velocity can be calculated by using Eq. (1.1).

(a)

(b)

Figure 1.1 Flow velocity measurement using cross-correlation

14

1.7 Background of Research Problem

In the Palm Oil Mill operation, there are two major wastes produced which

are liquid and solid wastes. The solid waste may contain palm Kernel shells,

mesocarp fibers as well as empty fruit bunches. The liquid waste produced from

palm oil extraction of a wet process primarily arises from oil room after separator or

decanter. When combined, this particular liquid waste and other wastes from

sterilizer condensate and cooling water is named the palm oil mill effluent (POME)

(Golder Associates, 2006).

A.L Ahmad and C.Y Chan defined POME as a high volume liquid waste

with a high tendency to pollute and it produces bad odours regardless of being non-

toxic and organic in nature. Due to this fact, the main concern would be on how to

tackle this problem so that there will be no considerable impact occurs since the palm

oil mill operation produces a lot of by-product and liquid wastes. The POME is

acidic in nature (pH 4-5), has discharged temperature of 80-90°C /50-60°C and non-

toxic (since no chemicals are added during extraction) (M. Ahmed, 2009). Most

recent palm oil mills required large treatment area and long treatment period of about

20 to 80 days since the biological treatment employed is conventional consisting of

anaerobic or facultative digestion (A.L. Ahmad, and C.Y. Chan, 2009).

The biological ponding system or the lagoon system treatment is building up

fast just as regular POME treatment system in Malaysia. This system consists of

dealing ponds, anaerobic, facultative and aerobic ponds. The ponding system usually

needs long retention time in excess of 20 days and the biogas is freed into the

atmosphere, where an average of 36% methane gas is released into the atmosphere

from an open tank digester. In Malaysia, the largest contributor to Green House Gas

(GHG) is the methane emission which comes from the palm oil industry. Carbon

emission credit can be gained for cutting down the emission of GHG (L.Y. Lang,

2007). Figure 1.2 shows the biological ponding system in palm oil milling.

15

Figure 1.2 Biological ponding system of Palm Oil Milling, Sime Darby Research

Sdn. Bhd. Carey Island, Banting Selangor, Malaysia, 2011.

The cutting down on the amount of palm oil mill effluent (POME) production

can definitely lessen the prior mentioned pollution. For the time being, there is no

proper system that has been created to observe the liquid waste percentage within the

vessel that makes it almost impossible to monitor the POME or waste matters such as

water, volatile matter, dirt or sludge inside the vessel until before it reaches the

separation process in the production. The only way that monitoring can be done so

far is to take the CPO sample out of the vessel during the production process to the

lab to be tested using biological method. The result of the composition can only be

received after 2 or 5 days.

Figure 1.3 (a) shows the flow diagram of palm oil milling process at Sime

Darby Research Sdn. Bhd. Carey Island and 2(b) shows the crude palm oil (CPO)

process.

16

(a) (b)

Figure 1.3 (a) Flow Diagram of Palm Oil Milling and (b) Crude Palm Oil Process -

Sime Darby Research Sdn Bhd, Carey Island, Banting Selangor, Malaysia, 2011.

The gravity-based separator technique is employed in recent work to isolate

the CPO and liquid wastes. The liquid waste produced from palm oil extraction of a

wet process mostly originates from the oil room. The CPO is required to be clarified

before is moved to the refining process right after the process of extraction. The main

reasons for the clarifying process is to isolate the crude palm oil from sludge or load

waste as shown in Figure 1.4.The light oil will stay at the top of the tank while the

heavy oil (mixed with liquid waste/sludge) will remain at the bottom as depicted in

Figure 1.4.

Figure 1.4 Clarifier Tank

17

The oil which stays at the top will flow to the purifier to yield clean CPO

while the remaining oil which is mixed with the sludge/liquid waste will undergo a

recovery process after it is channelled through a Decanter. The function of the

Decanter is to gather oil from the liquid waste. This is done before it goes to a

separation process which is to separate the oil from the sludge/liquid waste.

As an aid to such problem, the separation process monitoring to visualize the

percentage of liquid waste present within the vessel has become much easier with the

availability of instruments like the monitoring system. The collected data is very

useful in designing and creating better process equipment. Furthermore, the quality

of the CPO can be boosted due to the better control of certain process in aid of the

data which indirectly improve the POME treatment process. Today, our local palm

oil mill process has met new invention technology with the help of the tomography

technique. It is actually a research tool which has put aside the difficulties of

executing the measurements on the process plant.

1.8 Problem Statements

The ECT sensor design has been linked to many issues. There are only

several issues which were highlighted and discussed which recount to the

contribution in most of the previous research concerning to ECT as follows;

(i) It is hard to determine an inter-electrode capacitance which recount with the

relative permittivity distribution ε(r) and potential distribution φ(r), with the

use of the Laplace equation, it is yet too hard to be solved for the geometry

and the boundary conditions which therefore, in order to find the electric field

distributions, a finite element method (FEM) software simulation package is

so much needed. (W.Q Yang, 2010).

18

(ii) The sensor with external electrodes of the wall permittivity contributes to

non-linear changes in capacitance which somehow increase or decrease

relying on the thickness of the wall as well as the permittivity‟s of the sensor

wall and contents. (W.Q Yang, 2010).

(iii)The single-electrode excitation has the benefit of needing only one voltage

source, which can be substituted in sequence to the electrode being occupied

as a source. The setback of this method would be that, the sensitivity of the

sensor is higher in the wall area compared to the central area. This effect will

cause the mutual capacitance to be small, thus the electrode charges (and their

change) can also be very small (Gamio, J.C., 2002).

(iv) The sensitivity in different location between the electrode pair can differ

radically where the sensitivity distribution is not uniform. This is because of

the evident attribute of soft-field sensing (W.Q Yang, 2010).

(v) To select the distance between two planes is a trade-off. Poor cross

correlation will be achieved if the distance is too long due to the change of a

„tag‟ or pattern from the first plane to the second. On the otherhand, the time

resolution will be poor if the dsitance is too short. It is all due to the limited

number of samples for a „tag‟ or pattern to flow from the first plane to the

second plane. (W.Q Yang, 2010).

1.9 Research Objectives

The main objective of this research is to examine the capability of employing

a twin-plane ECT sensor with 16 portable electrodes with two different excitation

voltage source techniques so as to recognize the flow pattern, velocity profile as well

as the crude palm oil phase concentration to overcome the existing ECT system with

19

several enhancement and approaches. The specific objectives of the research are

listed as follows:

(i) To build a twin plane segmented ECT systems, non-invasive and nonintrusive

sensing module with two set of segmented electrodes mounted with one as

up-stream sensor while the other as a down-stream sensor.

(ii) To reconstruct the on line cross-sectional image of the distribution using

Linear Back Projection (LBP) method.

(iii)To display the cross-sectional image for up-stream & down-stream sensor

planes using Visual Basic programming platform in order to obtain the

information of flow from sensors. The information will then be analyzed and

be visualized in computer. The visualization results provide information

about the flow regime, velocity profile and the concentration distribution in

two-phase flow-rate measurement system incorporating with liquid flow

measuring device.

1.10 Research Scope

This research is divided into several parts, which are stated as follows:

i. Design a twin plane of ECT sensor with 16 electrodes every part of it is to

view the concentration profile image of gas and liquid. These images are

then cross-correlated on a pixel-by-pixel basis, i.e. the signals from each

pixel in image sensor plane 1 are cross-correlated with the signals from

their corresponding pixels in sensor plane 2 to obtain the velocity profile.

ii. Develop a 16 segmented portable sensor unit which is portable to be

assembled in different diameter sizes of pipeline, and it is flexible to

20

apply in any number due to different size of pipeline without the need of

redesigning the sensing module. The new approach of this sensing

module contains the integration intelligent electrode sensing circuit on

every each of electrode sensors. A microcontroller unit and a Data

Acquisition (DAQ) system have been incorporated in the electrode

sensing circuit and USB technology was applied into the data acquisition

system making the sensor able to work independently.

iii. Design a main controller circuit with new switching schemes using two

different transmitted signal excitation potentials/voltage source to a

different excited electrode pairs to yield a nearly uniform excitation field

across the sensor so as to examine the capacitance changes because of

different permittivity and to enhance the situation of non-linearization,

non-uniform potential distribution as well as less sensitivity in the ECT

measurement central area.

iv. Rebuild a cross sectional image in the usual and exact time with Visual

Basic 6 so as to verify the concentration profile of the flowing material

for both planes. This involves the employment of a linear back projection

algorithm in the rebuilding of the data with the use of an industrial

standard computer programming language.

1.11 Significant research contributions

The research contribution tries to enhance the situation of non-linear, non-

uniform potential distribution and less sensitivity in the ECT measurement central

area using a single excitation with two difference potential excitation /voltage source

technique. Develop a twin plane of 16-segmented sensor electrode for gas and liquid

flow velocity profile measurement in horizontal, downward inclined and upward

inclined flow and visualize the percentage of concentration distribution in two-phase

21

flow-rate measurement system for water, CPO (Crude Palm Oil) and sludge. Further

more in this research, the length of electrode design has been extend, in order

increase the electrode charges (and their change) in a range of 0.01 pF to 0.5pF.

The information gained will certainly be helpful in the work of monitoring

and governing the crude oil flow in a pipeline. This technique is resourcefully

utilized to observe the distribution of materials within the vessel and from this all

information regarding the flow mixture in conveying pipelines as well as the

concentration distribution will be identified from the cross sectional viewed. This

information is useful in designing better process equipment as well as to control

some processes to ensure that the quality of oil can be enhanced and thus minimizing

the waste from palm oil mills (POME). All journals that related to this research can

be referred in Appendix A.

CHAPTER 2

LITERATURE REVIEW

2.1 Research Background

Up until today, the production plants have shown that most flow-meters were

employed in measuring single-component materials such as petrol, gas, water and

many more. Nevertheless, there is a growing claim for multi-phase flow-meter due to

the effective use of resources and cost increasing request of effective processing

plants, while lessening and regulating pollution.

In today‟s, most processes have involved the utilization of various pipes or

vessels. They use mixture of products that are pumped along pipes (combination

product). For example, particulate materials (plastic, grain, and catalysts) are

disseminated along pipes by compressed air or pumped in liquid in manufacturing

process. However, the measurement of the amount delivered can only be done in one

way which is to isolate the components and then meter them individually or to fill

them into a tank or vessel then measure the volume or weight of that product in the

tank. The limitation is not inherent to the pumping mechanism; it is simply because

the technology for measuring the three-phase mixture flow-rate is not available. In

turn, the cost of the manufacturing operation (extra piping, valves, tanks and

weighing mechanism) will rise due to the deficiency of flow-meter technology for

these applications and resulting in possibilities of interference in the process flow.

23

For the oil and gas industry, the flow rate measurement in real time without

separating the phase is desirable in order to reduce cost, increase production and

reach excellence performance in oil and gas transport. For example, gas-oil is the

standard two-phase flows that can be easily discovered in chemical and petroleum

industries. Flow-rate measurement method of this type of two-phase flow has great

practical significance in reservoir management, process control, custody transfer

metering as well as fiscal metering. Yet, the inherent complexity of a two-phase

mixture makes it hard to determine the flow-rate (Z.Y. Huang et al, 2005).

2.2 Requirement Placed on Industrial Tomography

From its first beginnings, process tomography was aimed at providing

multiphase flow rate measurements as well as pictures of concentration distribution

(Salkeld J.A., 1991, Hayes, D.G.,1994). Measurement of flow rate through

tomography demands accurate images of concentration, high-speed data acquisition,

high-speed processing and some method of measuring velocity. Flow meters for

industrial use must also be robust and low-cost.

It makes possible for the chemical process engineer to indirectly observe the

chemical process that happens within the vessel which is hard to be done before. The

material‟s permittivity changes within the vessels are sensed by the capacitive

sensors installed. The permittivity distribution images are then developed based on

the collected data. This image is crucial to be used in understanding the nature of

activities that happens within the vessel by computing various key parameters

pertaining to the chemical process (S.M. Huang, et. al and C.G Xie et. al, 1992).

Apart from that, all the images may also assist in determining the future course of

control actions that must be taken if the need occurs. The technique has the

24

advantages of being non-intrusive in nature for measuring the capacitances, simple

transduction principle, fast response, simple and efficient algorithm applied to come

out with comparatively good resolution images. The sensors occupied are cost

effective capacitance sensors which is easy to build and handle.

In the mid-1980s, the work begins. It has somehow brought to the current

generation of Process Tomography systems which to create the phase or component

distribution image in an industrial process with the use of external sensors and is

deprived of triggering any perturbation to it. Example of appropriate processes are

those happening in stirring or mixing vessels, separators tanks or a pipeline carrying

multiphase flow and fluidized bed reactors. The University of Manchester (UMIST)

is a prominent university in industrial process tomography research area. This has

been proven since the late 1980‟s where they initiated a project on Electrical

Capacitance Tomography for imaging multi-component flows from oil wells (S.M.

Huang, et al., 1989). Fasching and Smith (1988) and Halow, J.S et al.(1990) claimed

that the Morgantown Energy Technology Center (METC), originated in the USA

designed a Capacitance Tomography system to measure the void distribution in gas

fluidized beds in approximately the same time. It is identified that the capacitance

transducers applied for both systems were only appropriate in an electrically non-

conducting situation (F. Wang, et al, 2009). Process Tomography is proclaimed to be

a cost effective technique ever since the development of low-cost parallel computers

during the 1980‟s. This development has helped to overcome the high cost and

slowness in image reconstruction using Van-Neuman computer architectures which

is seen as a problem.

In the early 90‟s, tomography system was applied in the field of industry.

Process tomography is the analogy of medical tomography. It carries the objective to

make available of visualizing the inner parts activities of industrial process like the

pneumatic pipe transportation activity. The designing of many devices like engines,

valve, boilers and turbo machines requires knowledge of fluid flow behaviour. These

flows experimental analysis puts the flow visualisation as the first step since the flow

226

REFERENCES

A Novel Tomographic Method of Measuring Flowrates of Mixtures in Pipes

Copyright (2002) Tomoflow ltd. http://www.tomoflow.com/tech.htm.

Olmos, A. M. (2008). Development of an Electrical Capacitance Tomography

system using four rotating electrodes. Journal Sensors and Actuators A:

Physical, Elsevier. Vol 148, 366-375.

Olmos, A. M. and Maroon, F. (2007). Simulation design of electrical capacitance

tomography sensors. IET Sci. Meas. Technology. 1(4), 216, 216-223.

A.L. Ahmad and C.Y Chan, (2009). Sustainability of palm oil Industries: An

Innovative Treatment via Membrane Technology. Journal of Applied

Sciences. 9 (17), 3074-3079.

A.J. Asis (2011). Processing & Engineering, Palm Oil Milling and Palm Kernel

processing. Sime Darby Research Sdn. Bhd. Carey Island, Banting Selangor.

A.W.S. Chia (2005). Poly-β-Hydroxybutyrate (PHB) Production From PalmOil Mill

Effluent (POME) Using Mixed Cultures Under Dynamically Transient Fed-

Batch System. Degree Thesis. Universiti Teknologi Malaysia , Skudai Johor.

Beck, M.S. and Plaskowski, A. (1987). Cross Correlation Flowmeters: Their Design

and Application.(Bristol: Adam Hilger).

C.K. Seong (2008). Electrical Capacitance Tomography (ECT) System with Mobile

Sensor for the liquid measurement. Master Thesis, Universiti Teknologi

Malaysia, Skudai Johor.

C.K. San (2002). Real time image reconstruction for fan beam optical tomography

system. Master Thesis. Universiti Teknologi Malaysia, Skudai Johor.

C.G. Xie, S.M. Huang, B.S. Hoyle, R. Thorn, C.P. Lenn, D. Snowden and Beck,

M.S. (1992). Electrical capacitance tomography for flow imaging: System for

227

development of image-reconstruction algorithms and design of primary

sensors. IEE Proc. G 139, 89–98.

D. Xie, Z. Huang, H. Ji, and H. Li. (2006). An Online Flow Pattern Identification

System for Gas–Oil Two-Phase Flow Using ElectricalCapacitance

Tomography. IEEE Transaction On Instrumentation and Measurement. Vol.

55, No. 5, 1833-1838.

E. Johana, R. Abdul Rahim, M.H. Fazalul Rahiman and S.Z. Mohd. Muji. (2012).

Design of Portable ECT for Crude Palm Oil Quality Monitoring System.

International Journal of Innovative Computing, Information and Control.

ISSN 1349-4198, Vol.8, No.1 (B).

E. Johana and R. Abdul Rahim. (2010). Multiphase Flow Reconstruction in Oil

Pipelines by Portable Capacitance Tomography. Proceedings of IEEE

Sensors, IEEE Sensors 2010 Conference. 1-4 November. Hawaii, USA, 273-

278.

Dickin, F.J., Hoyle, B.S., Hunt, A. and S.M. Huang. (1992). Tomographic imaging

of industrial process equipment: techniques and applications. IEEE Proc. G

139, No.1 .

Kuhn, .T., Schouten J.C., Mudde R.F., C.M van den Bleek. and Scarlett, B. (1996).

Analysis of chaos in fluidization using electrical capacitance tomography.

Meas .Sci. Technol, Institute of Physics Publishing. Vol.7,361–368.

F. Wang, Q. Marashdeh, L.S. Fan and Williams R.A. (2009). Electrical Capacitance,

Electrical Resistance, and Positron Emission Tomography Techniques and

Their Applications in Multi-Phase Flow Systems. International Journal of

Advances in Chemical Engineering, Science Direct. Vol.37, 179–222.

Chaplin G., Pugsley T. , Lani van der Lee, Kantzasand A. and Winters C. (2005).

The dynamic calibration of an electrical capacitance tomography sensor

applied to the fluidized bed drying of pharmaceutical granule. Meas. Sci.

Technol. Institute of Physics Publishing. Vol.16, 1281–1290.

Golder Associates (2006). Technology Overview-Palm Oil Waste Management. 05-

1113-244.

228

G.B. Zheng, N.D. Jin, X.H. Jia, P. July and X.B Liu (2008). Gas–liquid two phase

flow measurement method based on combination instrument of turbine

flowmeter and conductance sensor. International Journal of Multiphase

Flow, Science Direct. Vol.34, 1031–1047.

H. Li and Z. Huang (2000). Special Measurement Technology and

Application.Zhejiang University Press. Hangzhou.

H.G. Wang, T. Dyakowski, P. Senior, R.S. Raghavan andW.Q. Yang (2006).

Modelling of batch fluidised bed drying of pharmaceutical granules.”,

Chemical Engineering Science, Vol. 62, 1524 – 1535. Elsevier.

Eren, H. and Sandor, L.D. (2005). Fringe-Effect Capacitive Proximity Sensors for

Tamper Proof Enclosures. Sensors for Industry Conference.8-10 February.

Houston, Texas, USA,.

Halow, J.S., Fasching, G.E. and Nicoletti, P.(1990). Preliminary capacitance imaging

experiments of a fluidized bed. Advances in Fluidization Engineering, AIChE

Symposium Series. 276, Vol. 86, 41-50.

Hayes D.G. (1994). Tomographic flow measurement by combining component

distribution and velocity profile measurements in 2-phase oil/gas flows.PhD

Thesis. UMIST, UK.

H. Yan, F. Shao and S. Wang (1999), Simulation Study Of Capacitance Tomography

Sensors.1st World Congress on Industrial Tomography, 14-17 April. Buxton,

Greater Manchester.

I. Ismail, Gamio, J.C., S.F. Ahmed Bukhari and W.Q.Yang (2005). Tomography for

multi-phase flow measurement in the oil industry. Flow Measurement and

Instrumentation . Elsevier. Vol.16, 145-155.

Jaworski, J. and Dyakowski T. (2001). Application of electrical capacitance

tomography for measurement of gas-solids flow characteristics in a

pneumatic conveying system. Meas .Sci. Technology. Institute of Physics

Publishing.Vol. 12, 1109-1119.

Gamio, J.C. (2002). Acomparative analysis of single- and multiple-electrode

excitation methods in electrical capacitance tomography.

Meas.Sci.Technology. Vol.13,1799–1809. Institute of Physics Publishing.

229

Gamio, J.C., Castro, J., Rivera, L., Alamilla, J., Garcia-Nocetti, F. And Aguila, L.

(2005). Visualisation of gas–oil two-phase flows in pressurised pipes using

electrical capacitance tomography.Flow Measurement and Instrumentation.

Elsevier.Vol 16, 129–134.

Mirkowski, J. and Smolik, W.T. (2008). New Forward-Problem Solver Based on a

Capacitor-Mesh Model for Electrical Capacitance Tomography.IEEE

Transactions On Instrumentation And Measurements. Vol. 57, No. 5.

J. Frounchi, A.R. Bazzazi, K. Ebnabbasi and K. Hosseini (2003).High Speed

Capacitance Measuring System for Process Tomography.3rd World Congress

on Industrial Process Tomography. Banff, Canada.

Alme, K.J. and Mylvaganam, S. (2006). Analyzing 3D and Conductivity Effects in

Electrical Tomography System Using COMSOL Multiphysics EM Module.

Proceedings of the Nordic COMSOL Conference.

L.Y. Lang (2007). Treatability of Palm Oil Mill Efflueant (POME) using Black

Liquor in an Anaerobic Treatment Process.Msc.Thesis.Malaysia.

Loh W.W. (1998). Real-time monitoring of drilling cuttings transport using

electrical resistance tomography.PhD Thesis.UMIST

Beck, M.S. Bayars, M. and Dyakowski, T. (1997). Principles and industrial

applications of electrical capacitance tomography. Measurement & Control.

30 (7), 97–200.

M. Ahmed (2009). The Use of Microfillter Recovered Palm Oil Mill Enffluent

(POME) Sludge as Fish Feed Ingredient.Msc. Thesis. Malaysia.

Byars, M. (2001). Developments in Electrical Capacitance Tomography. Process

Tomography Limited. Cheshire, United Kingdom.

Flores, N., Gamio J.C. , Alemán, C.O. and Damián E. (2005). Sensor Modeling for

an Electrical Capacitance Tomography System Applied to Oil Industry.

Proceedings of the COMSOL Multiphysics User's Conference. Boston.

N.Mohd. Hasan and Azzopardi, B. J. (2007). Imaging stratifying liquid–liquid flow

by capacitance tomography. Flow Measurement and Instrumentation.

Elsevier. Vol 18, 241–246.

230

Patterson, R.P. and Z. Jie. (2003). Evaluation of an EIT reconstruction algorithm

using finite difference human thorax models as phantoms. Physiological

Measurement, Vol. 24, No. 2, 467.

P. Xue and L. Peng, (2010). A 36 Electrode Electrical Capacitance Tomography

Systems Using Asymmetric Electrode Strategy. World Congresson Industrial

Process Tomography (WCIPT6),6-9 September. Beijing, China.

Process Tomography Limited (2000). Calculation of Normalized Capacitances.

Application Note AN 5 Issue 1, Process Tomography Limited, Cheshire,

United Kingdom.

Process Tomography Limited (2001). Generation of ECT Images from Capacitance

Measurements. Application Note AN 1 Issue 3, March 2001, Process

Tomography Limited, Cheshire, United Kingdom.

Process Tomography Limited (2009). Fundamental Of ECT. Electrical Capacitance

Tomography System Operating Manual, Issue 1, Process Tomography

Limited, Cheshire, United Kingdom, December.

Waterfall, R .C., He, R., White, N.B. and Beck, C. M. (1995). Combustion imaging

from electrical impedance measurements. Meas .Sci. Technol., Vol.7, 369–

374.

White, R.B. and A. Zakhari. (1999). Internal Structures in Fluid Beds of Different

Scales: An Application of Electrical Capacitance Tomography. 1st World

Congress on Industrial Process Tomography. April 14-17. Buxton, Greater

Manchester,

Martin, R., Alemán C.O. and Castellanos A.R. (2005). Multiphase flow

reconstruction in oil pipelines by capacitance tomography using simulated

annealing. GeofísicaInternacional, Vol. 44, Num. 3, 241-250.

Williams, R.A. and Beck M.S. (1995). Introduction to process tomography: in

Frontiers in industrial process tomography, David M. Scott Eds., 1-7,

Butterworth-Heinemann Ltd.: Oxford.

R. Abdul Rahim, M.H. Fazalul Rahiman., Chan, K.S., Nawawi and S.W.

(2007).Non-invasiveimaging of liquid/gas flow using ultrasonic transmission-

231

model tomography.Sensors & Actuators: A. Physical. Elsevier.Vol.135, 337-

345.

S. Ibrahim, (2000). Measurement of gas bubles in a vertical water column using

optical tomography. Thesis of Doctor of philosophy Degree in Sheffield

Hallam University, United Kingdom.

S. Liu, Q.Chen, H.G.Wang, F.Jiang, I.Ismail and W.Q.Yang, (2005). Electrical

capacitance tomography for gas–solids flow measurement for circulating

fluidized beds. Flow Measurement and Instrumentation. Elsevier.Vol.16 ,

135–144.

S.M Huang, M., Liu, C. and Shen, L. C. (1995). Monitoring the contamination of

soils by measuring their conductivity in the laboratory using a contactless

probe. Geophysical Prospecting, Vol. 43, 759-778.

S.M Huang, Thorn, R., C.G Xie, Snowden, D. and Beck, M.S (1992) . Design of

sensor electronics for electrical capacitance tomography.IEE Proc. G,

Electron. Circuits & Sys.139,(1),pp. 83-88.

Salkeld, J.A. (1991). Process Tomography for the measurement and analysis of two-

phase oil-based flows.PhD Thesis, University of Manchester Institute of

Science and Technology (UMIST), UK.

Donthi, S.S. (2004). Capacitance based Tomography for Industrial Applications.M.

Tech. credit seminar report, Electronic Systems Group, EE Dept. IIT

Bombay.

Shepp, L.A., and Logan, B.F.(1974). The Fourier reconstruction of a head section.

.IEEE Trans. Nucl. Sci., Vol. 21, 21-43.

S.M. Huang,. Plaskowski, A.B, C.G. Xie and Beck, M.S.(1989). Tomographic

imaging of two component flow using capacitance sensors. Journal of

Physics E: Scientific Instruments, 22, 173–177.

S.F. Ahmed Bukhari and W.Q. Yang (2006). Multi-interface Level Sensors and New

Development in Monitoring and Control of Oil Separators.Sensor Journal,

MDPI, ISSN 1424-8220.

232

T.C. Wei (2003). Measurement of Two-Phase Imaging (Water/Gas Bubbles) By

Using Electrical Capacitance Tomography (ECT).Thesis of Degree.

Universiti Teknologi Malaysia, Skudai, Johor.

T.C. Tat (2003). Water/Oil Flowing Imaging of Electrical Capacitance Tomography

System. Thesis of Degree . Universiti Teknologi Malaysia, Skudai, Johor.

Thorn, R., S.M. Huang,, C.G., Xie, Salkeld, J.A., Hunt, A., and Beck,

M.S.(1990).Flow imaging for multi-component flow measurement. Flow

Measurement and Instrumentation. Elsevier. Vol.2, (2).

Dyakowski, T., York, T., Mikos, M., Vlaev, D. Mann, R., Follows G., Boxman, A.

and Wilson, M. (2000). Imaging nylon polymerisation processes by applying

electrical tomography. Chemical Engineering Journal. Elsevier. Vol.77, 105–

109.

Dyakowski T. (2000). Application of electrical tomography for gas-solids and liquid-

solids flows- a review.Powder Technology, Vol. 112,174-192. Elsevier.

York T. (2001). Status of electrical tomography in industrial applications. J.

Electron. Imaging, Vol. 10, no. 3, 608–619.

W. Q. Yang and York, T. A.,( 1999). New AC-based capacitance tomography

system. IEE Proc.- Sci. Meas. Technol., 146 (1), 47-53.

W. Q. Yang, Stott, A.L., Beck, M. S. and C. G. Xie, (1995b). Development of

capacitance tomographic imaging systems foroil pipeline measurements. Rev.

Sci. Instrum., 66 (8), 4326- 4332.

W.Q Yang (2010). Design of electrical capacitance tomography sensors. Meas .Sci.

Technol. Process Tomography Group, UMIST, Vol.21, 13.

W.Q Yang and L. Peng (2003). Image reconstruction algorithms forelectrical

capacitance tomography. Meas. Sci. Technol. Process Tomography Group,

UMIST Vol. 14 / R1–R13.

W.Q Yang and S. Liu (2000). Role of tomography in gas/solids flow

measurement.Flow Measurement and Instrumentation, Elsevier .Vol. 1, 237–

244..

233

W.Q Yang (1995a). Hardware Design of Electrical Capacitance Tomograohy

System.Meas .Sci. Technol., Institute of Physics Publishing.Vol.7, 225-232.

W.Q Yang, (2007). Tomographic Imaging based on Capacitance Measurement and

Industrial Applications. IEEE International Workshop on Imaging,Systems

and Techniques - IST 2007, May 4-5, Krakow, Poland.

W.Q. Yang, Chondronasios, A., Nattrass, S., Nguyen, V.T., Betting, M., I. Ismail and

McCann, H. (2004).Adaptive calibration of a capacitance tomography system

for imaging water droplet distribution.Flow Measurement and

Instrumentation, Elsevier.Vol.15 , 249–258.

W. Warsito and L.S. Fan (2001). Measurement of real-time flow structures in gas–

liquid and gas–liquid–solid flow systems using electrical

capacitancetomography (ECT). Chemical Engineering Science,

Pergamon.Vol 56, no 22, 6455–6462.

Wiegand, F., Hoyle and B.S.(1989). Simulations for parallel processing of

ultrasoundreflection-mode tomography with applications to two-phase flow

measurement. IEEE Trans. Ultrason.Ferroelectr. Freq. Control, Vol. 36,

652-60.

Y. Daoye, Z. Bin, X. Chuanlong, T. Guanghua and W. Shimin, (2009). Effect of

pipeline thickness on electrical capacitance tomography. Journal of Physics:

Conference Series 147, 012030.

L. Yang, Azzopardi, B.J., Baker, G., Belghazi, A. and Giddings, D. (2003). The

approach to stratification of a dispersed liquid–liquid flow at a sudden

expansion. 41st European two-phase flow group meeting.

W.Q.Yang, Spink, D.M., York, T.A. and McCann, H. (1999). An Image

ReconstructionAlgorithm Based on Landweber‟s Iteration Method for

Electrical Capacitance Tomography. Meas. Sci. Technol. Vol.10, 1065-1069.

Z. Cao, L. Xu and H. Wang, (2009). Image reconstruction technique of electrical

capacitance tomography for low-contrast dielectrics using Calderon's method.

Measurement Science and Technology, Vol.20, 104012-104027.

234

Z. Huang, B. Wang and H. Li,(2003). Application of electrical capacitance

tomography to the void fraction measurement of two-phase flow. IEEE

Transactions on Instrumentation and Measurement,52 (1), 7–12.

Z. Lin (2003). Gas–liquid Two-phase Flow and Boiling Heat Transfer. Xi’an

JiaotongUniversity Press, Xi‟an, China.

Z.Y. Huang, D. Xie, H. Zhang and H. Li.(2005). Gas-oil two-phase measurement

using an electrical capacitance tomography system and a Venturimeter. Flow

Measurement and Instrumentation, Elsevier.Vol.16, 177-182.