COMPREHENSIVE ACOUSTO-DIELECTROMETRIC NON-DESTRUCTIVE MONITORING OF POWER LINE DEVICES AND SUPPORTS USING BUILT-IN PERIPHERAL ARTIFICIAL INTELLIGENCE

Authors

C.Н. Верзунов Институт машиноведения, автоматики и геомеханики НАН КР

Keywords:

Non-destructive testing (NDT); Partial Discharge (PD); Gas-insulated switchgear; Composite and reinforced concrete transmission line supports; Acoustic emission and ultrasonic diagnostics; Dielectric spectroscopy; Bayesian data fusion; Edge computing; Machine learning and artificial intelligence; Predictive maintenance.

Abstract

The paper presents a systematic review of modern non-destructive testing (NDT) technologies for high-voltage objects, combining for the first time three approaches: acoustic emission, dielectric diagnostics, and embedded edge artificial intelligence (edge-AI). Examples of such objects include 10–45 kV gas-insulated switchgear (GISG) and composite/reinforced concrete transmission line (EL) supports. Data from peer-reviewed publications and several open signal sets (PD-Loc, VSB, etc.) are analyzed. Current processing algorithms are considered: convolutional neural networks (CNN), Bayesian fusion of multimodal data, and quantized models for edge-AI. The functions of leading industrial systems (EA UltraTEV, Rugged CPM601, OMICRON PARADIMO) are compared, and the architecture of an integrated acousto-dielectric device is proposed; it is shown that multimodal analysis can reduce the proportion of false alarms in partial discharge detection. The practical value of integrated monitoring for the transition to predictive maintenance of substations and transmission line supports, reducing sulfur hexafluoride (SF₆) leaks and increasing the mechanical reliability of structures is substantiated.

References

1. Samaitis V., Mažeika L., Jankauskas A., Rekuvienė R. Detection and Localization of Partial Discharge in Connectors of Air Power Lines by Means of Ultrasonic Measurements and Artificial Intelligence Models // Sensors. 2021. Vol. 21, № 1. P. 20. DOI: 10.3390/s21010020. URL: https://pmc.ncbi.nlm.nih.gov/articles/PMC7792959/ (дата обращения: 05.05.2025).

2. Blackman J., Averyt M., Taylor Z. SF₆ Leak Rates from High Voltage Circuit Breakers – U.S. EPA Investigates Potential Greenhouse Gas Emissions Source // Proc. 2006 IEEE Power Engineering Society General Meeting. Montréal, 18–22 Jun 2006. DOI: 10.1109/PES.2006.1709084. URL: https://www.epa.gov/system/files/documents/2022-05/leakrates_circuitbreakers.pdf (дата обращения: 05.05.2025).

3. EA Technology. UltraTEV® Detector²: Hand-Held Dual-Sensor Partial Discharge Detector [Электронный ресурс]. 2024. URL: https://eatechnology.com/solutions/partial-discharge-solutions/partial-discharge-detection/ultratev-detector/ (дата обращения: 05.05.2025).

4. Ohtsu M., Isoda T., Tomoda Y. Acoustic Emission Techniques Standardized for Concrete Structures // Journal of Acoustic Emission. 2007. Vol. 25. P. 21–28. URL: https://www.ndt.net/article/jae/papers/25-021.pdf (дата обращения: 05.05.2025).

5. Rugged Monitoring. CPM601-C Dual-Frequency Partial Discharge Monitor [Электронный ресурс]. 2024. URL: https://www.ruggedmonitoring.com/product/cpm601-c/ (дата обращения: 05.05.2025).

6. Jacobs L.J., Kurtis K.E., Sherman R.J., Burney D.C. Recommendations for Nondestructive Testing (NDT) of Concrete Components for Performance-Based Specifications. Final Rep. FHWA-GA-22-2015. Atlanta: Georgia DOT, 2022. 118 с. URL: https://rosap.ntl.bts.gov/view/dot/64475/dot_64475_DS1.pdf (дата обращения: 05.05.2025).

7. Michau G., Hsu C-C., Fink O. Interpretable Detection of Partial Discharge in Power Lines with Deep Learning // Sensors. 2021. Vol. 21, № 6. P. 2154. DOI: 10.3390/s21062154. URL: https://pmc.ncbi.nlm.nih.gov/articles/PMC8003486/ (дата обращения: 05.05.2025).

8. Chelmiah E.T., Madigan C.D., Kavanagh D.F. Partial Discharge – Localisation (PD-Loc) Dataset [Электронный ресурс]. IEEE DataPort, 2024. DOI: 10.21227/yy7g-2p79. URL: https://ieee-dataport.org/open-access/partial-discharge-localisation-pd-loc-dataset (дата обращения: 05.05.2025).

9. Jia Z., Fan S., Wang Z., Shao S., He D. Partial Discharge Defect Recognition Method of Switchgear Based on Cloud-Edge Collaborative Deep Learning // Scientific Reports. 2025. Vol. 15. Art. 10956. DOI: 10.1038/s41598-024-81478-9. URL: https://www.nature.com/articles/s41598-024-81478-9 (дата обращения: 05.05.2025).

10. VSB Power Line Fault Detection Dataset [Электронный ресурс]. Kaggle, 2019. URL: https://www.kaggle.com/competitions/vsb-power-line-fault-detection (дата обращения: 05.05.2025).

11. Partial Discharge Monitoring System // PAC World Magazine. 2024. Issue 68. URL: https://www.pacw.org/partial-discharge-monitoring-system (дата обращения: 05.05.2025).https://www.pacw.org/partial-discharge-monitoring-system

12. OMICRON Energy. PARADIMO 100 – Ultra-High-Frequency Partial Discharge Monitoring System for GIS and GIL [Электронный ресурс]. 2025. URL: https://www.omicronenergy.com/en/products/paradimo-100/ (дата обращения: 05.05.2025).