Fault Prediction for Power Transformer Using Optical Spectrum of Transformer Oil and Data Mining Analysis

Nur Afini Fauzi,Yang Sing Leong,Looe Hui Mun,Pin Jern Ker,N H Nik Ali,Md Zaini Bin Jamaludin,Ahmed H Sabry,Chin Kim Lo,Vimal Angela Thiviyanathan

doi:10.1109/access.2020.3011504

Nur Afini Fauzi, Yang Sing Leong + Show 7 more

Open Access

https://doi.org/10.1109/access.2020.3011504

Copy DOI

Abstract

Periodic preventive maintenance of power transformer should be conducted for its health monitoring and early fault detection. Transformer oil is a vital element where its contents and properties need to be monitored during the service life of a power transformer. This paper presents an optical spectroscopy measurement from 200 nm to 3300 nm to characterize the transformer oil, which were sampled from the main tanks and ‘on-load tap changer’ of power transformers. The correlation of the optical characteristics in the range of 2120 nm to 2220 nm to the Dissolved Gas Analysis results and Duval Triangle interpretation demonstrates that the low energy electrical discharges, high energy electrical discharges as well as the thermal faults rated at temperatures above 700°C in power transformers can be accurately predicted. For faster and accurate analysis of fault prediction, a data mining analytics tool was constructed using Rapid Miner server to analyze and verify the predictions for a total of 108 oil samples. For the optimization, continuous iterations were performed to determine the best absorbance-wavelength combination that can improve the accuracy of the prediction. The performance of the optical spectroscopy technique integrated with data analytic tool was analyzed and it was found that the technique contributes to a high accuracy of 98.1% in fault prediction. It is a cost-effective and quicker complementing approach to carry out pre-screening of the transformer oil in order to know the condition of the power transformers based on the transformer oil’s optical characteristics.

Highlights

There are various power transformer condition monitoring techniques such as frequency response analysis [1]–[3], partial discharge analysis [4]–[8] and Dissolved Gas Analysis (DGA) [9], [10] that have been employed in determining the condition of power transformers
This research focuses on the optical characterization of transformer oil in relation to the fault conditions based on DGA results and Duval’s triangle interpretation
The findings indicate that samples with D2 and D1 fault conditions exhibit high optical absorbance peaks near 2172 nm

Summary

INTRODUCTION

There are various power transformer condition monitoring techniques such as frequency response analysis [1]–[3], partial discharge analysis [4]–[8] and Dissolved Gas Analysis (DGA) [9], [10] that have been employed in determining the condition of power transformers. For DGA applications, infrared (IR) spectroscopy [18], [19] has been used to determine the absorption properties of dissolved gases in transformer oil. Raman [25], [26] and near-infrared (NIR) spectroscopy [27] have demonstrated good progress in characterizing the optical absorption properties of dissolved gases in transformer oil. This paper proposes to categorize the fault condition directly based on the absorbance spectrum of the transformer oil measured in the ultraviolet-visible-near infrared (UV-Vis-NIR) region without determining the concentrations of the fault gases. This work proposes a procedure to incorporate data from the optical spectra of transformer oil into the Rapid Miner analysis tool for fault prediction in power transformer. VOLUME 8, 2020 value as well as the range of the wavelength to categorize and predict the fault condition based on the absorbance spectrum of transformer oils

EXPERIMENTAL DETAILS

OPTICAL SPECTROSCOPY MEASUREMENT

DATA MINING ANALYTIC TOOL

RESULTS AND DISCUSSION

CONCLUSION