Instrumentation & Measurement Magazine 24-2 - 107

controlled lab conditions that lead to achieving high classification accuracy. However, the real challenge is to apply
this technology under field conditions. It is expected that
several practical problems will arise under field conditions such as external noise, uncontrolled defects and
impact of other factors that may have not been considered
during training the ML model.
◗◗ Application of deep learning instead of conventional ML
algorithms: While application of ML has achieved high
classification accuracy (at least under controlled conditions), extracting the features is usually a tedious process.
It requires significant trial and error to reach the optimum (or semi-optimum) features. Recently, using deep
learning algorithms like convolutional neural network
(CNN) showed great potential to replace the existing
conventional ML algorithms. One of the major benefits
of using deep learning algorithms is their ability to learn
directly from the input data without the need to extract
any features. Several attempts have been recently applied
to detect damages in both ceramic and non-ceramic insulators with promising results [18], [19].
◗◗ Integrate ML algorithms with drones: Due to their efficiency, utility companies start to use manually controlled
drones extensively in their transmission line inspection
instead of using helicopters. Integrating ML algorithms
can further improve the drones' inspection capability as
demonstrated with recently published work [20]. So far,
the main sensor that has been used with drones in the
context of overhead line inspection is regular cameras.
Integrating other sensors like acoustic sensors, IR
cameras can further improve the inspection accuracy. For
example, using both regular and IR cameras improved
the ability of the ML algorithm to estimate the insulator pollution condition [21]. Furthermore, using acoustic
sensors along with IR cameras enhanced the ability to
detect faulty insulators [22].

[4]	 R. S. Gorur, D. Shaffner, W. Clark and R. Vinson, " Utilities share
their insulator field experience, " Transmission and Distribution
World, vol. 57, no. 4, pp. 17-27, 2005.
[5]	 G. H Vaillancourt, J. P Bellerive, M. St-Jean and C. Jean. " New live
line tester for porcelain suspension insulators on high-voltage
power lines, " IEEE Trans. Power Delivery, vol. 9, no. 1, pp. 208-219,
1994.
[6]	 S. Anjum, S. Jayaram, A. El-Hag and A. Naderian, " Detection and
classification of defects in ceramic insulators using RF antenna, "
IEEE Trans. Dielectrics and Electrical Insulation, vol. 24, issue 1, pp.
183-190, Feb. 2017.
[7]	 S. Polisetty, A. El-Hag and S. Jayaram, " Classification of common
discharges in outdoor insulation using acoustic signals and
artificial neural network, " IET High Voltage, vol. 4, no. 4, pp. 333338, 2019.
[8]	 S. Yi, L. Qin, T. Liangrui, Y. Qiuxia, " A detection algorithm of
hydrophobic levels based on triangle module operator, " in Proc. 7th
Int. Conf. Fuzzy Systems and Knowledge Discovery, pp. 22-26, 2010.
[9]	 S. Yi, L. Qin, T. Liangrui, Y. Qiuxia, " The fusion estimating
algorithm of hydrophobic level based on D-S evidence theory, " in
Proc. 3rd IEEE Int. Conf. Broadband Network and Multimedia Technol.
(IC-BNMT), pp. 860-864, 2010.
[10]	T. Tokoro, S. Yanagihara and M. Kosaki, " Diagnosis of
hydrophobic condition of silicone rubber using dielectric
measurement and image analysis, " in Proc. Conf. Electrical
Insulation and Dielectric Phenomena, pp. 281-284, 2005.
[11]	L. Maraaba, Z. Al-Hamouz and H. Al-Duwaish, " A neural
network-based estimation of the level of contamination on highvoltage porcelain and glass insulators, " Electrical Eng., 2017.
[12]	R. de Aquino, J. Bezerra, M. Lira, G. Santos, O. Neto and C. de
O.Lira, " Combining artificial neural network for diagnosing
polluted insulators, " in Proc. Int. Conf. Neural Networks (IJCNN),
pp. 179-183, Jun. 2009.
[13]	X. Jianyuan, S. Wei, C. Rong, T. Yun and L. Xin, " A new
combination forecasting model for ESDD prediction of
suspension porcelain insulators, " in Proc. 1st Ann. Int. Conf. Electric
Power Equipment - Switching Technology (ICEPE-ST), pp. 279-282,

Conclusions

Oct. 2011.

In this article, the potential of applying ML in assessing the
health conditions of outdoor insulators is highlighted. A
summary of current research activities in this area is briefly
introduced and the shortcoming of the existence research is
briefly described.

[14]	M. A. Salam, S. M. Al-Alawi, A. A. Maqrashi, " Prediction of
equivalent salt deposit density of contaminated glass plates using
artificial neural networks, " J. Electrostatics, Elsevier Science, vol.
66, issue 9-10, pp. 526-530, Sep. 2008.
[15]	J. Li, C. Sun, W. Sima, Q. Yang and Jianlin Hu, " Contamination
level prediction of insulators based on the characteristics of

References

leakage current, " IEEE Trans. Power Delivery, vol. 25, no. 1, pp.

[1]	 N. Bhatia, A. H. El-Hag and K. Shaban, " Machine learning-based
regression and classification models for oil assessment of power

417-424, Jan. 2010.
[16]	A. K. Abouzeid, A. El-Hag and K. Assaleh, " Equivalent salt

transformers, " in Proc. IEEE Int. Conf. Informatics, IoT, and Enabling

deposit density prediction of silicone rubber insulators under

Technol. (ICIoT'20), pp. 400-403, Feb. 2020.

simulated pollution conditions, " Electric Power Components and

[2]	 L. Khalayli, H. Al Sagban, H. Shoman, K. Assaleh and A. ElHag, " Automatic inspection of outdoor insulators using image

Syst., vol. 46, no. 10, pp. 1121-1131, Dec. 2018.
[17]	I. Patel, I. Maarouf, S. Soltan, A. Saad, A. Al-Taher, A. El-Hag

processing and intelligent techniques, " in Proc. IEEE Electrical

and K. Assaleh, " Image processing based estimation of ceramic

Insulation Conf. (EIC 2013), Jun. 2013.

insulator pollution levels, " in Proc. 5th Electric Power and Energy

[3]	 A. A. Soltani and A. El-Hag, " Denoising of radio frequency partial
discharge signals using artificial neural network. " Energies, vol.
12, no. 18, pp. 1-14, 2019.
April 2021	

Conversion Sys. (EPECS 18), Apr. 2018.
[18]	G. Kang, S. Gao, L. Yu and D. Zhang, " Deep architecture for
high-speed railway insulator surface defect detection: denoising

IEEE Instrumentation & Measurement Magazine	107
Instrumentation & Measurement Magazine 24-2

Table of Contents for the Digital Edition of Instrumentation & Measurement Magazine 24-2
