1 of 22

NanoCommunication-based Impermeable Region Mapping for Oil Reservoir Exploration

Liuyi Jin†, Lihua Zuo ‡ , Zhipei Yan†, Radu Stoleru†�

†Department of Computer Science & Engineering, Texas A&M University

‡ Department of Mathematics, Texas A&M University-Kingsville

​

NanoCom 2019

​

lenss.cse.tamu.edu

2 of 22

Outline

  • Introduction
    • Motivation
    • State of the art and challenge
    • Our contribution
  • Problem Formulation
    • System model
    • Impermeable area geometry characterization problem
  • Proposed Solution
    • Heuristic well placement algorithm
  • Evaluation
    • Simulation implementation and performance evaluation
  • Conclusion and Future Work

2

lenss.cse.tamu.edu

3 of 22

Motivation

  • Global energy requirement will increase 1.5 times over the next 2 decades (Kong, X. and Ohadi, M., 2010). Oil still plays a significant role in providing worldwide energy.

Figure source: BP energy outlook 2019

3

lenss.cse.tamu.edu

4 of 22

Motivation

  • Trillions of dollars of investment in oil will be needed
    • We want to improve the efficiency and reduce the cost of extracting oil from underground reservoir

Figure source: BP energy outlook 2019

4

lenss.cse.tamu.edu

5 of 22

Motivation

  • Reservoir characterization – physical properties of reservoir: geometry, net thickness, permeability, porosity…
  • Impermeable area characterization

Figure source: http://www.longbeach.gov/energyresources/about-us/oil/subsidence/

5

lenss.cse.tamu.edu

6 of 22

State of the Art

  • Impermeable area characterization:
  • Britto (Britto and Sageev SPE California Regional
  • Meeting 1987) and Nestor (Nestor and Heber, SPE Conference Paper 1983): transient analysis method to detect impermeable area
  • Søndergaard (Søndergaard and Auken, SEG Annual
  • Meeting 2008): data integration to map the large-scale aquifer
  • Nanotechnology advancements:
  • Nishtha (Nishtha 2017, dissertation, Characterisation of skin-based thz communication channel for nano-scale body-centric wireless networks): THz-based network architecture under human skin to monitor human health
  • Challenges:
  • Current solutions are costly and impractical
  • No real time information can be obtained

​

6

lenss.cse.tamu.edu

7 of 22

Our Contributions

  1. Underground oil exploration through nanodevices
  2. optimal impermeable area mapping problem formulation
  3. An accurate mapping of impermeable areas is possible
  4. For future research, we propose to employ more sophisticated cyber-physical components

7

lenss.cse.tamu.edu

8 of 22

Problem Formulation

 

8

lenss.cse.tamu.edu

9 of 22

Problem Formulation

 

(Akkas GLOBECOM’10, Terahertz channel modeling of underground sensor networks in oil reservoirs)

9

lenss.cse.tamu.edu

10 of 22

Problem Formulation

Impermeable Area Geometry Characterization Problem

​

Pick the “tightest” streamline loop

There are 2 challenges to overcome:

  1. Increase the number of well pairs leads to huge cost(approx. $500 per meter of well depth).
  2. the shape of the streamline is hard to compute (What we can exactly know is only the length of each single streamline): arc is used to approximate streamline shape

10

lenss.cse.tamu.edu

11 of 22

Problem Formulation

Impermeable Area Geometry Characterization Problem

​

 

 

11

lenss.cse.tamu.edu

12 of 22

Problem Formulation

Impermeable Area Geometry Characterization Problem

​

 

 

 

 

12

lenss.cse.tamu.edu

13 of 22

Problem Formulation

Impermeable Area Geometry Characterization Problem

​

 

13

lenss.cse.tamu.edu

14 of 22

Proposed Solution

 

 

14

lenss.cse.tamu.edu

15 of 22

Performance Evaluation

  • Qualitative Results

Algorithm 1 decision of which well-pair to include contributes to a more and more accurate mapping of the impermeable square area (from (b) through (f))

15

lenss.cse.tamu.edu

16 of 22

Performance Evaluation

  • Quantitative Results – area mapping precision

 

16

lenss.cse.tamu.edu

17 of 22

Performance Evaluation

  • Quantitative Results – location correctness

 

17

lenss.cse.tamu.edu

18 of 22

Performance Evaluation

  • Quantitative Results – efficiency

 

18

lenss.cse.tamu.edu

19 of 22

Conclusions

  1. Map underground impermeable areas using streamline approximations.
  2. Intersection of arc pairs as our prediction of the impermeable region
  3. Perpendicular searching method to minimize number of well pairs

19

lenss.cse.tamu.edu

20 of 22

Future Work

  • Concave impermeable region shapes: If the impermeable region is a concave shape, our solution will only obtain the convex hull of the unknown region.
  • We are only using the distance between two adjacent nanodevices. If we employ the distance among non sequential nanodevices, we have the chance to reconstruct the shape of the streamline.
    • This will make our mapping more accurate.

20

lenss.cse.tamu.edu

21 of 22

Questions?

21

lenss.cse.tamu.edu

22 of 22

Proposed Solution

Streamline Simulation

  1. Eclipse reservoir simulator produces the pressure and velocity distribution
  2. Pressure and velocity distributions are used to produce streamlines

The simulator results are in macroscale. To cope with this, we discretized the real streamlines with approximated line segments with nanoscale length

22

lenss.cse.tamu.edu