Keywords: heavy-oil thermal recovery, detection of sand level, CCD. 1. Introduction. Heavy oil takes advantage of proportion in world petroleum resources.
Available online at www.sciencedirect.com
Physics Procedia 25 (2012) 1823 – 1828
2012 International Conference on Solid State Devices and Materials Science
Technology of Sand Level Detection Based on CCD Images Bing Jianga, Xianglin Lia, Xiaohui Chena, Tengfei Zhanga, Chi Fenga,Fei Zhangb b
a College of Automation Nanjing University of Posts and Telecommunications Jiangsu Province, China College of Electronics and Information Jiangsu University of Science and Technology Jiangsu Province, China
Abstract Heavy oil takes advantage of proportion in world petroleum resources. Thermal recovery technology, the chief means of heavy-oil exploitation, has been widely applied in development of world heavy-oil reservoir. To study the effect of sand-control technology in the process of heavy-oil thermal recovery, A HTRSTS (Heavy-oil Thermal Recovery Simulation Testing System) has been build. The detection of sand level in sand container is very important. The sand level detection technology adopted in this system is image processing technique based on CCD. Sand container image is taken by CCD, and then HTRSTS locks the interface between sand and liquid through CCD scanning. The preliminary experimental result shows that the standard deviation is about 0.02 liter, which could satisfy practical requirement quite well.
© 2011 ofof [name 2012 Published Published by by Elsevier Elsevier Ltd. B.V.Selection Selectionand/or and/orpeer-review peer-reviewunder underresponsibility responsibility Garryorganizer] Lee Open access under CC BY-NC-ND license. Keywords: heavy-oil thermal recovery, detection of sand level, CCD
1.
Introduction
Heavy oil takes advantage of proportion in world petroleum resources. The recoverable reserves of heavy oil exceed that of ordinary oil.[1-4] Therefore heavy oil’s effective production has always been a key problem facing world petroleum industry. Compared with light oil, heavy oil could exert stronger drag force on stratum sand under the same flow velocity condition because of its higher viscosity. So shear failure is easier to occur in sand body, which led to sand production.[5-7] On the other hand, the washout effect caused by high pressure difference and injected hot stream in thermal-recovery process also reinforce heavy oil sand production. Traditional high-cost sand control technology produce little effect on Silt and fine sand control, therefore its application in thermal recovery well has been severely circumscribed.[8-11] To study the effect of sand-control technology in the process of heavy-oil thermal recovery, A HTRSTS has been build. In this system, the detection of sand level in sand container is very important. The dynamic detection results from different sand-control modules are directly used to evaluate the sand-control effect. The sand level detection technology adopted is image processing
1875-3892 © 2012 Published by Elsevier B.V. Selection and/or peer-review under responsibility of Garry Lee Open access under CC BY-NC-ND license. doi:10.1016/j.phpro.2012.03.317
1824
Bing Jiang et al. / Physics Procedia 25 (2012) 1823 – 1828
technique based on CCD. Preliminary experimental results show that the system could satisfy practical requirement quite well. 2.
HTRSTS
HTRSTS uses SCADA (Supervisory Control and Data Acquisition) architecture. An IPC (Industrial PC) is used to implement the following functions: Ł Gathering data from different kinds of sensors such as pressure sensors, temperature sensors and so on. ł Data analysis and processing. ŃResult display and false alarm. Siemens S7-300 PLC realizes detection and control of motors. Fig. 1 shows the system flow chart.[12-13] Water in WP or liquor in LT is sent into LA through pipe 2 and pipe 3 by metering pump 1. Gear pump stirs liquid in LA to mix well. This liquor mix stage should be first prepared. The following four processes could be performed independently or sequentially according to different modules and different demands: Ł liquid in LA is sent into sand agitator through pipe 6 by metering pump 2, and then sent into modules by mud pump. Meanwhile, water in WP is sent into desand cyclone through pipe 1 by metering pump 1 for sand-removing. ł liquid in LA is sent into modules directly through pipe 5 by metering pump 2. Meanwhile, water in WP is sent into desand cyclone through pipe 1 by metering pump 1 for sand-removing. Ń liquid in LT is sent into modules though pipe 2, pipe 4 and pipe 5 by metering pump 1. ń liquid in LT is sent into sand agitator through pipe 2, pipe 4 and pipe 6 by metering pump 1, and then sent into modules by mud pump. HTRSTS evaluates the effect of sand-control module according to dynamic sand-level data. 3.
The sand level detection technology based on CCD
CCD is used to take sand container image, and than HTRSTS locks the interface between sand and liquid via CCD scanning. IPC calculates sand content according to the distance CCD moving. The sand level detection set is shown in Fig. 2. To avoid the influence of ambient light, a fluorescent lamp provides back light for the whole detection set, meanwhile, the frosted glass before fluorescent lamp improves the uniformity of back light. Sand container is a cylindrical vessel made of quartz glass with a diameter of 100mm. In order to prevent the possible bodily injury caused by the vessel explosion, a metal shield surrounds the sand container. Two slits symmetrically locate at left and right sides of metal shield for the purpose of 30 32
6
1
29
9
20
28
31 11
2
22
4
27
10
21
13 14
5
23
15 16 17 18
3
7
8
25 19
12
26
24
Bing Jiang et al. / Physics Procedia 25 (2012) 1823 – 1828 1. sand box 2. Weight-measure unit 3. PLC 4. sand agitator 5. mud pump 6. flow meter 7. IPC 8. output unit 9. metering pump2 10.LA( liquid agitator) 11. gear pump 12. UPS 13. multi-layer well module 14. Fracturing and Sand-Control module 15. Compound Perforation module 16. Gravel packs module 17. horizontal well module 18. Vertical well module 19. pressure and temperature sensors 20. metering pump1 21.WP(water pool) 22. LT(liquid tank) 23. desand cyclone 24. filter 25. CCD 26. sand container 27. pipe 1 28. pipe 2 29. pipe 3 30. pipe 4 31. pipe 5 32. pipe 6 Fig. 1 6ystem flow chart
The images of slit taken by CCD are shown in Fig. 3 and Fig. 4. Fig. 3 illustrates that interface has not been detected and CCD keeps scanning. Fig. 4 indicates that the interface has been locked. 2
3 5 4
C
1
1. fluorescent lamp 2. frosted glass 3.sand container 4. slit 5. CCD Fig. 2 Sand level detection set
Fig. 5 shows the gray value of line 3 and line 296 when interface is locked. Edge data are discarded for avoiding influence of invalid data. It can be seen that the data in sand sand-level detection, which are 10mm in width. CCD scans the slit up and down driven by step motor. IPC locks the interface between sand and liquor by image processing.[14-19] area and that in liquid area have obviously difference, which is the basis for interface judgment.
Fig. 3 The slit image when interface has not been detected
1825
1826
Bing Jiang et al. / Physics Procedia 25 (2012) 1823 – 1828
In the process of scanning, images are captured every second. At first LAViΰ Line Average Valueαof valid data in slit area should be calculated, then TGV(Top gray value) and BGV(Bottom gray value) can be expressed as 8
296
¦ LAV
i
TGV = i=292
¦ LAV
i
ΰ1α
BGV =
5
i=3
ΰ2α
5
Fig.4 The slit image when interface has been detected Gray value of line 3 and line 296 300
250
Gray value
200
150
100 line 3 line 296
50
0
0
50
100
150
200 colum
250
300
350
400
Fig.5 Gray value of line 3 and line 296 when interface is locked
TGV and BGV are used for interface judgment. Fig. 6 shows LAVi when interface is locked and unlocked separately. Average gray value 300
locked unlocked
Average gray value
250
200
150
100
50
0
0
50
100
150 line
200
250
Fig. 6 LAVi when interface is locked and unlocked
300
Bing Jiang et al. / Physics Procedia 25 (2012) 1823 – 1828
If the interface enters the visual field of CCD according to judgment, the interface line number could be obtained from: TGV BGV ΰ3α Min i ( LAVi ) 2 CCD Camera locates the center of its view field at interface, and then IPC calculates the distance CCD moving, which is sand-level height. So the sand content could be worked out. Preliminary experiments of sand level detection were carried out. Considering the original location of sand container and CCD, initial value of sand content is 0.47L. Test result is shown in Table 1. Table 1 Sand Content Test Result (Unit: liter) Real sand content 0.5 1 1.5 2 2.5 3 3.5 4 4.5
Detection Result 0.48 0.98 1.48 2.01 2.47 3.02 3.53 3.99 4.48
Test result shows that the standard deviation is about 0.02 liter. 4.
Conclusions
To study the effect of sand-control technology in the process of heavy-oil thermal recovery, a HTRSTS (Heavy-oil Thermal Recovery Simulation Testing System) has been developed. In this system, the sand level detection technology adopted is image processing technique based on CCD. Sand container image is taken by CCD. The data in sand area and that in liquid area have obviously difference, which is the basis for interface judgment. HTRSTS locks the interface via CCD scanning. Preliminary experiments of sand level detection have been carried out. Result shows that the sand content standard deviation is about 0.02 liter, which could satisfy practical requirement quite well.
5.
Acknowledgments
This work was supported by the National Natural Science Foundation of China (51007007, 60805039, 61001077), Scientific Research Project of Office of Education in Jiangsu Province (11KJB130002, 08KJB510015) and Scientific Research Foundation for the Introduction of Talents (NY207146ΕNY209019).
References [1]
L. Zhang, GH. Que and WN. Deng, “Dynamic study on the dispersion of water-soluble catalysts for heavy-oil
hydrocracking,” Journal of Petroleum Science and Engineering, vol. 73, no. 1, pp. 27-32, August 2010. [2]
SD. Razavi and R. Kharrat, “Application of Cyclic Steam Stimulation by Horizontal Wells in Iranian Heavy Oil
Reservoirs,” Scienia Iranica Transaction C-chemistry and Chemical Engineering, vol. 16, no. 2, pp. 125-139, 2009.
1827
1828 [3]
Bing Jiang et al. / Physics Procedia 25 (2012) 1823 – 1828 G. Butler, et al, “Maximize liquid yield from extra heavy oil Next-generation hydrocracking processes increase
conversion of residues,” Hydrocarbon Processing, vol. 88, no. 9, pp. 51-55, 2009. [4]
S. Moghadam, M. Nobakht and YG. Gu, “Theoretical and physical modeling of a solvent vapour extraction (VAPEX)
process for heavy oil recovery,” Journal of Petroleum Science and Engineering, vol. 65, no. 1, pp. 93-104, 2009. [5]
Q. Jiang, et al, “Review of Thermal Recovery Technologies for the Clearwater and Lower Grand Rapids Formations in
Cold Lake, Alberta,” Journal of Canadian Petroleum Technology, vol. 49, no. 9, pp. 57-68, 2010. [6]
ZF. Li, “Casing cementing with half warm-up for thermal recovery wells,” Journal of Petroleum Science and
Engineering, vol. 61, no. 2, pp. 94-98, 2008. [7]
HF. Thimm and K. Kwasniewski, “Prediction of scales in boilers for thermal recovery projects,” Journal of Canadian
Petroleum Technology, vol. 47, no. 1, pp. 26-28, 2008. [8]
D. Underdown and H. Chan, “Evaluation of Sand-Control Completions in the Duri Steamflood, Sumatra, Indonesia,” SPE
Drilling & Completion, vol. 24, no. 1, pp. 137-143, 2009. [9]
Q. Dasheng and P. Bailin, “Physical Experiment Research of Mechanical Sand Control for Tarim Donghe Oilfield,”
Petroleum Science and Technology, vol. 27, no. 9, pp. 952-958, 2009. [10]
RD. Pourciau, “Deepwater extended-reach sand-control completions and interventions,” SPE Drilling & Completion, vol.
22, no. 2, pp. 157-164, 2007. [11]
C. McPhee, C. Farrow and P. McCurdy, “Challenging convention in sand control: Southern North Sea examples,” SPE
Production & Operations, vol. 22, no. 2, pp. 223-230, 2007. [12]
DA. Phan, and DC. Truong, “Component-based Design for SCADA Architecture,” International Journal of Control
Automation and Systems, vol. 8, no. 5, pp. 1141-1147, 2007. [13]
ZH. Sun, and XM. Tian, “SCADA in Oilfields,” Measurement & control, vol. 43, no. 6, pp. 176-178, 2010.
[14]
Q. Li and B. Zhang, “Template matching based on image gray value,” Visual Communications and Image Processing,
vol. 5960, pp. 614-622, 2005. [15]
B. Rieger, et al, “On curvature estimation of ISO surfaces in 3D gray-value images and the computation of shape
descriptors,” IEEE Transaction on Pattern Analysis and Machine Intelligence, vol. 26, no. 8, pp. 1088-1094, 2010. [16]
N. Dekker, LS. Ploeger and M. VanHerk, “Evaluation of cost functions for gray value matching of two-dimensional
images in radiotherapy,” Medical Physics, vol. 30, no. 5, pp. 778-784, 2003. [17]
CS. Lin, et al, “A positioning model of a two CCD camera coordinate system with an alternate-four-matrix look-up table
algorithm,” Optics and Lasers in Engineering, vol. 48, no. 12, pp. 1193-1199, 2010. [18]
JJ. Dong and XJ. Zhao, “High Speed Data Acquisition System Based on ARM & Linear-CCD,” ISTM/2009: 8th
International Symposium on Test and Measurement, VOLS 1-6 pp. 3115-3118, 2009. [19]
XY. Ma, CH, Rao and HQ, Zheng, “Error analysis of CCD-based point source centroid computation under the
background light,” Optics Express, vol. 17, no. 10, pp. 8525-8541, 2009.
