Condition Monitoring of High Voltage Transformers Using Spectrophotometric, Dissolved Gas Analysis, and Acoustic Emission Techniques

Abstract

High voltage transformers are the vital and costly equipment in the entire power system network. The failure of such equipment incur huge financial loss as well as power supply disruption to the consumers. Therefore, continuous monitoring of the high voltage transformers is an inevitable task, particularly as they are always under the influence of electrical and thermal stress which lead to the incipient faults such as partial discharge, arcing, pyrolysis etc. In this proposed work, the spectroscopy techniques have been used such as Ultraviolet-visible (UV-visible), Fourier Transform Infrared (FTIR), and Nuclear Magnetic Resonance (NMR) techniques to monitor the condition of the transformer as they have a lot of advantages like they are non-intrusive, have high throughput, sensitive, and do not harm the samples. As mineral oil has certain drawbacks like poor biodegradability, low flash point, and future shortage, this work also has been focused on vegetable oils such as rapeseed oil (MDEL eN 1204), coconut oil, and soybean oil. UV-visible spectroscopy has the potential to identify the age of the thermally affected-degraded oil samples both mineral and vegetable oils by measuring their concentrations in terms of peaks arose in the spectra at different wavelengths. Moreover, the obtained results with UV-visible spectroscopy demonstrated that it could qualitatively analyze the samples by visually showing the degradation. FTIR spectroscopy is a capable technique to identify the various functional groups present in the thermally degraded oil samples. Particularly, it has been instrumental to identify the aromatic, carbonyl, and acidic groups present in the degraded sample oils. Further, investigation of the samples with NMR spectroscopy provided the picture of the hydro-carbon structure in degraded oil samples thus determining the physical and chemical information. Also, to monitor the high voltage transformers, the well-known dissolved gas analysis (DGA) technique has been used. Because of some limitations of the individual DGA diagnostic methods such as key gas, roger’s ratio, IEC gas ratio, Doernenberg ratio, Duval triangle and as no single method could evaluate the transformer condition accurately, in this proposed work, they have been integrated to form into a single system using LabVIEW software and obtained the better results. Finally, as the partial discharge (PD) is no small problem in the power transformers, acoustic emission technique has been used to detect and locate the PD occurrence spot so as to take the fault preventive measures.