DETERMINING SATURATION USING ELECTRICAL IMPEDANCE TOMOGRAPHY EIT

1
DETERMINING SATURATION USING ELECTRICAL IMPEDANCE
TOMOGRAPHY (EIT)
Robert W. Stacey, Kewen Li and Roland N.
Horne Stanford Geothermal Program rstacey_at_stanfor
d.edu
2
Motivation

• Residual saturation is critical in vapor
  dominated fields
• Improve energy recovery and resource development
• Safely determine resource connectivity

The Geysers CA
3
Motivation

• Saturation is very difficult to measure
• Previous research with X-Ray CT scan
• Metal core holders are required due to high
  pressure and temperature

X-Ray CT Scan Saturation Measurements
4
EIT Theory

• Known
• I, applied current
• Vn, measured voltage
• Unknown
• s impedance distribution
• Problem, nonlinearity OO(s)
• Voltage distribution function of impedance

5
EIT Lab Setup
PC
Switch Controller
Current Generator
Core
6
Electrode Design
5cm
10cm

• Berea sandstone core
• 48 electrodes, 3 rings of 16
• Attached with conductive epoxy
• 0.05mA applied current

7
Measuring Potential Field
V1
Bottom Ring
Top Ring
Middle Ring

1
2
16
15
3
14
4
Top Ring
5
13
12
6
11
7
8
10
9
-
Current electrode measurements omitted due to
skin effect
8
Data Acquisition System

• 48 measurements per excitation
• 2,304 measurements total
• 12 second scan time

9
Data Processing Software

• EIDORS Electrical Impedance Tomography and
  Diffuse Optical Tomography Reconstruction
• Started in 2002, Polydorides EE PhD
• Group goal is to promote collaboration between
  medical and industrial research groups in EIT and
  optical tomography

Nick Polydorides and William R B Lionheart 2002 A
Matlab toolkit for three-dimensional electrical
impedance tomography a contribution to the
Electrical Impedance and Diffuse Optical
Reconstruction Software project Meas. Sci.
Technol. 13(12)1871-1883
10
Data Processing

• MATlab based software
• Finite element model for forward calculations
• Regularized non-linear solvers for obtaining a
  unique and stable inverse solution
• Graphical Software

Nick Polydorides and William R B Lionheart 2002 A
Matlab toolkit for three-dimensional electrical
impedance tomography a contribution to the
Electrical Impedance and Diffuse Optical
Reconstruction Software project Meas. Sci.
Technol. 13(12)1871-1883
11
Image Reconstruction
II)
Initial saturation, distilled water
I)
Water, lower ring
High Conductivity Low Conductivity
IV)
III)
Water, middle ring
Water, upper ring
12
Real-Time Imaging Experiment

• Imbibed distilled water overnight
• Saline drip displacement
• 194 Scans
• Reconstruction time 20 minutes

13
Real-Time Imaging Experiment
Saline Displacement
14
Real-Time Imaging Experiment
Middle Ring
Bottom Ring
Vertical Slice
Top Ring
15
EIT Calibration

• Determined Resistivity-Saturation Relationship

16
EIT Calibration

• Archies Law
• Modification added due to fluid resistance
  dependent upon saturation

17
EIT Calibration

• Resistivity Index vs. Saturation (DC system)

18
EIT Calibration

• Resistivity Index vs. Saturation (AC system 40
  Hz)

19
EIT Calibration

• Resistivity Index vs. Saturation (AC system 40
  Hz)

20
EIT Calibration

• Comparison 5_13 (AC) and 4_6 (DC)

21
Conclusions

• EIT is capable of imaging saturation fronts in
  real-time
• Numerical dispersion currently makes the reliable
  detection of sharp saturation fronts difficult.
• EIT is capable of measuring the saturation
  distribution at saturations accurately between
  20 lt Sw lt 65 in a rock core.
• The DC current causes erroneous resistivity
  results at high saturations (Sw gt 65) due to ion
  mobility
• The modification to Archies equation was
  critical in capturing the physics of the system
• EIT is an inexpensive, practical, compact, and a
  safe alternative to the CT scan

22
Future Work

• Implementing high frequency AC current source to
  negate ion mobility effects
• Further quantification of results with mass
  balance and X-ray CT scan
• Performing a sensitivity analysis on the data
  collection and inversion software
• Expanding development into fractured geothermal
  cores
• Investigating field applications

23
Acknowledgments

• This research was conducted with financial
  support to the Stanford Geothermal Program from
  the US Department of Energy under grant
  DE-FG07-02ID14418, the contribution of which is
  gratefully acknowledged.

24
Questions?