Page 113 - IJEEE-2023-Vol19-ISSUE-1
P. 113
Abed, Wali, & Alaziz | 109
(b) Scattering plot examination. Electr. Commun. Comput. Eng., pp. 1–4, 2020. doi:
10.1109/ICECCE49384.2020.9179393.
Fig. 15: Continued. [3] W. R. Chalgham, A. C. Seibi, and M. Mokhtari,
“Simulation of Sound Wave Propagation Inside a
V. CONCLUSION Spherical Ball Submerged in a Pipeline,” Proc. 2016
The results of a numerical examination of the velocity, COMSOL Conf. Bost., p. 2016, 2016.
pressure, and temperature distribution have been [4] M. M. Hathal, “Oil fouling in double - pipe heat
satisfactorily modeled over various velocities and kinds of exchanger under liquid - liquid dispersion and the
fluid. In the case of oil, the production of the most dead influence of copper oxide nanofluid,” Heat Transfer, vol.
zones is seen, whereas, in the case of gas, the values are at 48, Issue 5, pp. 1963-1981, July 2019. doi:
their lowest. Due to the thermally poor qualities of the gas 10.1002/htj.21473.
phase, insulation activity is encouraged to take place within [5] J. V. N. de Sousa, C. H. Sodré, A. G. B. de Lima, and S.
the pipe (thermal conductivity). The 2.5 meters per second R. de F. Neto, “Numerical Analysis of Heavy Oil-Water
velocity enables more fluid mixing in the region where the Flow and Leak Detection in Vertical Pipeline,” Adv. Chem.
temperature distribution is most uniform. Compared to water Eng. Sci., vol. 03, no. 01, pp. 9–15, 2013. doi:
and gas, oil has a higher maximum pressure drop because of 10.4236/aces.2013.31002.
its viscous forces and dead zones. The highest pressure drop [6] C. W. Liu, Y. X. Li, Y. K. Yan, J. T. Fu, and Y. Q.
and highest velocity are both provided by the maximum Zhang, “A new leak location method based on leakage
velocity. In this paper, An SVM was proposed for detecting acoustic waves for oil and gas pipelines,” J. Loss Prev.
leakage at four points in 3 m, 4.5 m, 6 m, and 9 m pipelines Process Ind., vol. 35, pp. 236–246, May 2015. doi:
using a pressure sensor. The dataset of this study was 10.1016/j.jlp.2015.05.006.
obtained from the simulation with three parameters of [7] S. I. Kam, “Mechanistic modeling of pipeline leak
velocity, pressure, and temperature. Each parameter consists detection at fixed inlet rate,” J. Pet. Sci. Eng., vol. 70, no.
of three values (0.1, 1, 2.5 m/s). The SVM and DT models 3–4, pp. 145–156, Feb. 2010. doi:
get an average accuracy (validation) of 98.8%, and 99.87%, 10.1016/J.PETROL.2009.09.008.
respectively for classification. Comparing the results [8] S. Valizadeh, B. Moshiri, and K. Salahshoor,
obtained, we conclude that we obtain higher accuracy when “Multiphase pipeline leak detection based on fuzzy
training and testing the DT. So, for this data type, it is best to classification,” AIP Conf. Proc., vol. 1159, no. 2009, pp.
use the DT algorithm because it is faster and more accurate 72–80, 2009. doi: 10.1063/1.3223958.
than another algorithm. For future work, different [9] L. Dong, S. Chai, and B. Zhang, “Leak detection and
classification algorithms could be applied and the resulting localization of gas pipeline system based on wavelet
classification accuracy could be compared to achieve higher analysis,” Proc. 2nd Int. Conf. Intell. Control Inf. Process.
efficiency. ICICIP 2011, no. PART 1, pp. 478–483, 2011. doi:
10.1109/ICICIP.2011.6008290.
CONFLICT OF INTEREST [10] R. Hu, H. Ye, G. Wang, and C. Lu, “Leak detection in
The authors have no conflict of relevant interest to this pipelines based on PCA,” 2004 8th Int. Conf. Control.
article. Autom. Robot. Vis., vol. 3, pp. 1985–1989, 2004. doi:
10.1109/ICARCV.2004.1469466.
REFERENCES [11] C. Verde, “Multi-leak detection and isolation in fluid
[1] R. Kroes, L. Amundsen, and R. Hoffmann, “Crude pipelines,” Control Eng. Pract., vol. 9, no. 6, pp. 673–682,
Jun. 2001. doi: 10.1016/S0967-0661(01)00026-0.
oil-water flow in horizontal pipes,” 2013. [12] M. Francis, A. Ross, and A. James, “Experimental
http://essay.utwente.nl/69533/ Observation And Analysis Of Inverse Transient For
[2] W. A. Wali, “Carbon Dioxide Conversion Control Pipeline Leak Detection,” ASCE, vol. 133, no. March, pp.
Based on Microwave Plasma Technology,” Int. Conf. 519–530, 2014.
[13] Y. Liu, X. Ma, Y. Li, Y. Tie, Y. Zhang, and J. Gao,
“Water pipeline leakage detection based on machine
learning and wireless sensor networks,” Sensors
(Switzerland), vol. 19, no. 23, pp. 1–21, 2019. doi:
10.3390/s19235086.
[14] U. Baroudi, A. A. Al-Roubaiey, and A. Devendiran,
“Pipeline Leak Detection Systems and Data Fusion: A
Survey,” IEEE Access, vol. 7, pp. 97426–97439, 2019. doi:
10.1109/ACCESS.2019.2928487.
[15] Y. Song and S. Li, “Gas leak detection in galvanised
steel pipe with internal flow noise using convolutional
neural network,” Process Saf. Environ. Prot., vol. 146, pp.
736–744, 2021. doi: 10.1016/j.psep.2020.11.053.
[16] A. Terfous, A. Hazzab, and A. Ghenaim, “Predicting
the drag coefficient and settling velocity of spherical