Noor Dzulaikha Daud, Nur Laili Suhaimi and Mohamad Kamarol Mohd Jamil*

1 School of Electrical & Electronic Engineering, Universiti Sains Malaysia, Nibong Tebal, Penang, Malaysia ndzulaikha@usm.my, lailisuhaimi950@gmail.com, eekamarol@usm.my

Abstract: UHF loop antenna sensors have been widely used for detecting partial discharges (PD) in high-voltage (HV) systems, their application in detecting surface discharges (SD) remains underexplored. Current technologies primarily focus on detecting bulk discharge events, leaving a significant gap in understanding the effectiveness of loop antennas for surface discharge detection. Surface discharges exhibit distinct electromagnetic characteristics that require optimized detection methods. This study investigates the use of loop antennas as Ultra High Frequency (UHF) sensors for detecting surface discharges on glass-type outdoor insulators. It delves into the design, development, and evaluation of the loop antenna, focusing on its performance in identifying dominant frequencies associated with SD events. Experimental results demonstrate the antenna's effectiveness at its resonant frequencies and its capability to detect SD over varying distances. These findings suggest that loop antennas are highly sensitive and cost-effective solutions for real-time monitoring of SD activities, improving maintenance strategies and increasing the reliability of high-voltage (HV) power transmission systems.

Keywords: Surface discharge; high voltage; UHF sensor; leakage current; insulator

1. Introduction

Power systems rely on effective insulation to ensure the consistent and safe transmission of electrical energy [1]. Outdoor insulators are critical to the reliability of power networks. However, surface discharges (SD) on these insulators can gradually affect the material, resulting in probable failures and decreased reliability. Timely detection of SD is essential to prevent unexpected equipment failures and improving the overall safety and durability of high-voltage (HV) systems.

Environmental factors, such as pollution and moisture, can lead to the formation of surface discharge on insulators [2, 3]. These surface discharges can degrade the insulating material, leading to increased risk of insulation failure and flashover, which can disrupt power transmission. Over time, surface discharge can cause significant damage to insulators, reducing their efficiency and lifespan. There has been extensive research into the performance of outdoor high voltage insulator in high polluted environments.

SD poses a significant threat to the insulation system's performance and longevity. SD is characterized by localized breakdowns occurring at or near the surface of the insulating material. It is often associated with leakage currents that result from the accumulation of contaminants on the insulator's surface [4].

Conventional methods for detecting SD on outdoor insulators have limitations in terms of sensitivity and real-time monitoring capabilities [5, 6]. The IEC 60270:2000 defines standard conventional method for detecting PD which requires an electrical connection through a coupling capacitor connected in parallel with the monitored equipment [7]. These methods are restricted by large physical dimensions for high voltage applications and impracticality for continuous monitoring due to their invasive nature [8].

UHF signals are known to be generated during PD events [9]. Research has discussed on the development of UHF sensor for detecting PD in high voltage equipment focusing on various UHF antennas and their suitability for PD diagnostics [10-12]. Factors such as detection range, directional characteristics, and ease of fabrication are considered. The UHF technique utilizes the electromagnetic signals emitted during surface discharge events with frequencies ranging from hundreds of megahertz to gigahertz and can be captured using suitable sensors and

Received: April 10th, 2025. Accepted: June 7th, 2025

DOI: 10.15676/ijeei.2025.17.2.7

instrumentation [13]. This method offers several advantages, including real-time monitoring, enabling prompt detection of incipient discharge events [14]. Additionally, the non-intrusive nature of the UHF monitoring minimizes the impact on the insulator's normal operation [15, 16]. This research contributes to the development of effective diagnostic tools for the proactive management of outdoor insulators, helping to prevent catastrophic failures and improving the reliability of power systems[17, 18]. A comparative analysis of shunt resistor, IMC and loop antenna techniques under varying voltages conditions is presented, providing insights by quantitatively correlating frequency shifts and bandwidth variations to discharge intensity and sensor sensitivity to improve UHF-based SD detection accuracy.

2. Methodology

A. Development of loop antenna as UHF sensor

Antenna is used as a sensor to detect EM waves radiated by the discharge source. Fig. 1 shows the completed wired loop antenna. The materials required for this construction of wired loop antenna include a 1-meter length of coaxial cable (such as RG-58), a ferrite core for the balun, enamel-coated copper wire suitable for high-frequency applications, a soldering iron and solder, heat shrink tubing, BNC connector, and wire strippers and cutters.

Figure 1. A wired loop antenna

For a loop antenna, the resonant frequency is related to the circumference of the loop, which is the total length of the wire forming the loop. The fundamental resonant frequency occurs when the circumference of the loop is approximately equal to one wavelength of the signal. The circumference, C of a circular loop is related to its diameter, D by the formula \(C = \pi D\). Therefore, the resonant frequency can be expressed in terms of the diameter.

Given:

Circumference of loop antenna, C = 1 m The relationship between the circumference, C=1 m and the diameter, D of a circle is given by:

\[C = \pi D\] \[D = \frac{c}{\pi}\] \[D = 0.318m\] (1)

Given:

• Diameter of loop antenna, D = 0.318 m
• Speed of light, \(c \approx 3 \times 10^8\) meters per second. The resonant frequency, f is given by:

\[f = \frac{c}{\pi D}\] \[f = \frac{3 \times 10^8}{\pi \times 0.318}\] (3)

\[f \approx 3 \times 10^8 \approx 300 \, MHz\]

Therefore, the resonant frequency of a loop antenna with a diameter of 0.318 m is approximately 300 MHz.

B. Evaluation on frequency range of developed loop antenna

Fig. 2 illustrates the measured S11 parameters (magnitude in dB) of a loop antenna design obtained using a network analyzer. The S11 parameter, also known as the reflection coefficient, reveals how much power is reflected back from the antenna, a critical factor for determining its impedance matching and overall efficiency. The plot displays the amplitude of the S11 parameter across a frequency range from 10 MHz to 1 GHz.

Figure 2. S11 results from measurement of loop antenna

From the results of S11 parameters shown in Fig. 2, the deepest dip on the graph is at 946 MHz, indicating the frequency at which the antenna has the best impedance match. At this frequency, reflection is minimized, suggesting that most of the power is transmitted rather than reflected, making 946 MHz the optimal operating frequency for this loop antenna design. SD events that occur at this frequency can be effectively detected by the antenna.

SD events typically generate electromagnetic emissions that span a range of frequencies. An antenna operating at its resonant frequency is most sensitive to signals at that frequency. Therefore, if the resonant frequency of the loop antenna aligns with the frequency spectrum of SD events, the antenna will efficiently detect these events. Furthermore, the presence of multiple resonant frequencies enhances the antenna's ability to cover a broader frequency spectrum, making it suitable for detecting a wider range of SD events and their harmonics. The network analyzer's measurements illustrate the antenna's performance characteristics across the specified frequency range, demonstrating its capability to detect surface discharge events at these optimal frequencies.

C. Verification of frequency transmission-loop antenna (receiver) and S11 parameter analysis

A signal generator was connected to the transmitter antenna to inject a specified frequency. The transmitter antenna then transmitted this frequency to the receiver antenna, which was positioned at a measured distance. In this project, a loop antenna was used as the receiver antenna. The loop antenna received the transmitted signal and was connected to a signal analyzer to verify the received frequency. Fig. 3 depicts the setup for frequency injection from transmitter to receiver antenna.

The verification process involved comparing the frequency value displayed on the signal analyzer with the injected frequency from the signal generator. A match between these frequencies confirmed that the loop antenna successfully received the transmitted frequency.

Figure 3. Setup for frequency injection from transmitter to receiver antenna

D. Experimental setup

Fig. 4(a) illustrates the schematic circuit diagram of the experimental setup, whereas Fig. 4(b) depicts the SD experiment configuration within the HV laboratory. A 240V, 50Hz input voltage was fed to a voltage regulator, with a circuit breaker in between to protect against overload orshort circuit. A 5kVA step-up transformer is used to increase the voltage source. The voltage regulator controlled the voltage supplied to the transformer. In the experiment, the applied voltage was 4.5kV and a 33kΩ current limiting resistor was used to keep the current below 0.1A, limiting potential short circuit currents [19, 20] .

Figure 4. (a) Schematic circuit diagram and (b) SD experiment in HV laboratory

A capacitive voltage divider, with a ratio of 1000:1, scale down the high voltage signals for safe measurement by an oscilloscope, which has a voltage input limit of 400V. The high voltage from the transformer's secondary winding was connected to the pin end of the glass insulators, designed to support high voltage cables. The cap end of the insulator was connected to the grounding system for safety, providing a controlled path to the ground for any fault current or induced voltages.

A NaCl solution with an ESDD level of 0.5 mg/cm² was used to create leakage current on the insulator by spraying it on the surface of glass insulator based on IEC 60507:2013 [21]. The shunt resistor and IMC are used as SD detection devices, measured by an oscilloscope and a spectrum analyzer, to observe the SD events. These methods serve as reference points for detecting SD events with the loop antenna. The antenna is connected with spectrum analyzer to record SD events emitted due to the occurrence of leakage current on glass insulator.

3. Results and discussion

Fig. 5(a)-(c) shows the frequency spectrum of SD events when tested with shunt resistor, IMC, and loop antenna at 5 kV, 13 kV and 17 kV applied voltages of respectively. Numerically, the loop antenna recorded the highest amplitude signals, with dominant frequencies shifting from approximately 255 MHz at 5 kV to 900 MHz at 17 kV, increasing by a factor of 3.5. In comparison, the shunt resistor and IMC serve as reference points for the loop antenna, showed frequency increases with voltage, but the loop antenna demonstrated greater sensitivity to higher frequency signals generated by more intense SD. The SD spectra results shown in the figures after the filtration of background noise.

Fig. 6 illustrates the frequency bandwidth at 17 kV applied voltage, comparing the calculated frequencies with the measured frequencies of SD events on polluted insulator. Based on Fig. 6, the frequency of 11 MHz is the only value that remains consistent across both calculated and measured frequency bandwidth. Additionally, the SD spectra at 11 MHz, 255 MHz, and 900 MHz were consistently detected by all SD measurement devices; shunt resistor, IMC and loop antenna. Nevertheless, these frequencies; 255 MHz and 900 MHz, were not detected by oscilloscope through rise time measurement. This is likely due to the small magnitude of SD events at these frequencies. Despite the discrepancy between measured and calculated value, the consistency of these frequencies across various measurement techniques confirms their accuracy as dominant frequencies of SD events.

Figure 5. (a)-(c): SD spectra at 5kV, 13kV, 17kV applied voltage

Figure 6. Frequency bandwidth at 17kV applied voltage

On the other hand, the calculated frequencies, while present in the spectrum, does not stand out as a dominant frequency. This is because, despite its association with high-magnitude SD events as observed with the oscilloscope, it occurs less frequently compared to lower- magnitude events. When analyzed with a spectrum analyzer, this frequency appears as a subtle signal rather than a prominent one, yet it remains relevant for a thorough understanding of SD events. Thus, the dominant frequencies of SD events are 11 MHz, 255 MHz, and 900 MHz.

Besides, those frequencies detected by loop antenna are close to its resonant frequencies, making it more effective at detecting SD events that occur around these resonant frequencies. Furthermore, the loop antenna's performance over a distance of 2 meters was also measured. We observe that even with the increased distance between the glass insulator and the loop antenna, the signal strength of SD events remained detectable. The antenna's ability to detect numerous SD events at its resonant frequencies, even over an increased distance, suggests that it is highly sensitive.

4. Conclusion

The optimized loop antenna shape significantly influences frequency response characteristics such as bandwidth, return loss and resonant frequency, enhancing SD detection across a broad frequency range and at extended distances through integration with multiple sensor techniques. The results show that the dominant frequencies of the SD spectra are 11 MHz, 255 MHz, and 900 MHz. Additionally, SD spectra tested with the loop antenna, which was simultaneously connected with the IMC and shunt resistor, not only captured a wide range of frequencies but also showed SD events radiated very effectively at the antenna's resonant frequency. The antenna continued to detect surface discharge (SD) events effectively at 2 meters, with minimal reduction in signal strength compared to measurements taken at 1 meter.

5. Acknowledgment

Appreciation and gratitude are to be given to the authority of the School of Electrical and Electronics Engineering University Sains Malaysia for providing the materials and guidance to complete the project. The research and implementation of the project would never be completed without the facilities provided by the school.

6. References

[1]. V. Pradeep, K. Baskaran, and S. I. Evangeline, "An improved transfer learning model for detection of insulator defects in power transmission lines," Neural Computing and Applications, vol. 37, no. 9, pp. 6951-6976, 2025/03/01 2025, doi: 10.1007/s00521- 025-11011-0.

• [2]. Z. Zhang, F. Yang, H. Zhang, C. Zhou, Y. Li, and H. Liu, "A probabilistic neural network assessment method for insulator pollution level based on discharge noise," Measurement, vol. 242, p. 115869, 2025/01/01/ 2025, doi: https://doi.org/10.1016/j.measurement.2024.115869.
• [3]. L. Maraaba, K. Al-Soufi, T. Ssennoga, A. M. Memon, M. Y. Worku, and L. M. Alhems, "Contamination level monitoring techniques for high-voltage insulators: A review," Energies, vol. 15, no. 20, p. 7656, 2022.
• [4]. G. Wang, W. Zhao, W. Tian, Y. Zhao, W. Yuan, and P. Yan, "Characterizing the Contamination Degree of Bore Insulators in Solid-Armature Launchers Using Partial Surface Conductivity," IEEE Transactions on Plasma Science, pp. 1-8, 2025, doi: 10.1109/TPS.2025.3548680.
• [5]. R. V. de Andrade Lira et al., "Elliptical UHF sensor for partial discharge detection," Sensors and Actuators A: Physical, vol. 348, p. 113981, 2022.
• [6]. S. K. Azam, M. Othman, H. A. Illias, T. A. Latef, M. T. Islam, and M. F. Ain, "Ultrahigh frequency printable antennas for partial discharge diagnostics in high voltage equipment," Alexandria Engineering Journal, vol. 64, pp. 709-729, 2023.
• [7]. High-voltage test techniques: partial discharge measurements, I. Standard, 2000.
• [8]. G. A. Hussain, W. Hassan, F. Mahmood, M. Shafiq, H. Rehman, and J. A. Kay, "Review on partial discharge diagnostic techniques for high voltage equipment in power systems," IEEE Access, vol. 11, pp. 51382-51394, 2023.
• [9]. T. Arunkumar, H. M. Wilson, T. P. Pandit, and S. J. Lee, "Electrical power generation and utilization in advanced desalination systems," Renewable and Sustainable Energy Reviews, vol. 210, p. 115211, 2025/03/01/ 2025, doi: <a href="https://doi.org/10.1016/j.rser.2024.115211">https://doi.org/10.1016/j.rser.2024.115211</a>.
• [10]. M. A. Asmat and R. Abd Rahman, "Partial Discharge Detection using UHF Sensor on High Voltage Equipment," Evolution in Electrical and Electronic Engineering, vol. 5, no. 2, pp. 246-253, 2024.
• [11]. B. Habib, O. Al Zaabi, N. Harid, K. Al Hosani, and M. Alkhatib, "Condition monitoring based on partial discharge diagnostics using UHF sensors: A comprehensive state-of-the-art review," IEEE Transactions on Dielectrics and Electrical Insulation, 2024.
• [12]. M. Taher et al., "Conformal and Flexible Antennas in Ultra-High Frequencies: Prospects and Challenges for Partial Discharge Diagnostics," IEEE Access, 2025.
• [13]. J. A. Ardila-Rey, D. Figueroa, F. P. Torres, S. Govindarajan, B. A. De Castro, and R. Schurch, "Bioinspired Ultra High Frequency Antenna for Partial Discharge Detection in High-Voltage Equipment," IEEE Transactions on Instrumentation and Measurement, 2024.
• [14]. S. Asefi et al., "Review of High Voltage Instrument Transformer Condition Monitoring," IEEE Transactions on Dielectrics and Electrical Insulation, 2024.
• [15]. G. Wu, T. Zhang, B. Cao, K. Liu, K. Chen, and G. Gao, "A Review and Progress of Insulation Fault Diagnosis for Cable Using Partial Discharge Approach," IEEE Transactions on Dielectrics and Electrical Insulation, 2024.
• [16]. H. Kumar, M. Shafiq, K. Kauhaniemi, and M. Elmusrati, "A review on the classification of partial discharges in medium-voltage cables: detection, feature extraction, artificial intelligence-based classification, and optimization techniques," Energies, vol. 17, no. 5, p. 1142, 2024.
• [17]. A. H. M. Ibrahim, S. K. Sadanandan, T. Ghaoud, V. S. Rajkumar, and M. Sharma, "Incipient fault detection in power distribution networks: review, analysis, challenges and future directions," IEEE Access, 2024.
• [18]. A. Mohanty et al., "Power system resilience and strategies for a sustainable infrastructure: A review," Alexandria Engineering Journal, vol. 105, pp. 261-279, 2024.

• [19]. J. Gehring, R. Schwanninger, A. Nowak, B. Wunder, V. Lorentz, and M. März, "Method for Detection and Limitation of Short-Circuit Currents for Semiconductor Circuit Breakers in LVDC Grids Using a Pre-Saturated Inductor," IEEE Open Journal of Power Electronics, vol. 6, pp. 524-536, 2025, doi: 10.1109/OJPEL.2025.3555797.
• [20]. J. Liu et al., "Study on Impedance Characteristics of 10kV Magnetic Biased Superconducting Fault Current Limiter," Singapore, 2025: Springer Nature Singapore, in The Proceedings of the 11th Frontier Academic Forum of Electrical Engineering (FAFEE2024), pp. 13-29.
• [21]. Artificial pollution tests on high-voltage ceramic and glass insulators to be used on A.C. systems, I. Standard, 2013.

Noor Dzulaikha Daud earned a Bachelor's degree in Electrical System Engineering from Universiti Malaysia Perlis. She earned both her Master's degree in Electrical Power Engineering, in 2013 and her Ph.D. in Electrical Engineering in 2022 from Universiti Teknologi Malaysia. Currently, she is a senior lecturer at School of Electrical & Electronics, Universiti Sains Malaysia. She has authored and co-authored several journal and conference papers. Her research focuses on Micromachining, Electrical Power System and Renewable Energy. She can be contacted via

email at ndzulaikha@usm.my.

Nur Laili Suhaimi graduated with a Bachelor of Engineering (Honours) IN Electrical Engineering from Universiti Sains Malaysia (USM) in 2024. She completed an internship at Tenaga Nasional Berhad (TNB), where she gained practical experience in monitoring, analyzing, and troubleshooting power distribution networks, including high-voltage substations and transformers, ensuring reliable energy delivery and system efficiency. She currently works as an Electrical Engineer at INV New Material Technology where she has been actively involved in the setup process, gaining handson experience and a solid understanding of key engineering operations.

Mohamad Kamarol Mohd Jamil (Senior Member, IEEE) received the B.Eng. degree (Hons.) in electrical engineering from Universiti Technology MARA, Malaysia, in 2000, and the M.Eng. and D.Eng. degrees from the Kyushu Institute of Technology, Japan, in 2005 and 2008, respectively. In 2002, he joined as a University ASTS Fellow with Universiti Sains Malaysia (USM), where he was a Senior Lecturer, in 2008, an Associate Professor, in 2014, and was promoted to Professor in Dec 2024. He was a Senior Engineer with Sankyo Seiki (M) Sdn. Bhd., for

almost eight years. He was a Visiting Researcher with the High Voltage Laboratory, Kyushu Institute of Technology, from 2013 to 2014, and the Chiba Institute of Technology, Japan, in February 2020. His research interests include the insulation properties in oil palm, solid dielectric material, insulation properties of environmentally benign gas, and PD detection technique for insulation diagnosis of power apparatus and electrical machine. He is also involved in temperature rise and short-circuit electromagnetic study of bus bar systems and HVDC systems. Professor Ir. Dr. Mohamad Kamarol is a Professional Engineer and a member of the Board of Engineers Malaysia and the Institution of Engineers Malaysia. He received the Chatterton Young Investigator Award from the IEEE International Symposium Discharges and Electrical Insulation in Vacuum, in 2006.