Vibroacoustic transformer condition monitoring

Starkey, Dean (2016) Vibroacoustic transformer condition monitoring. [USQ Project]

[img]
Preview
Text (Main Project)
Starkey_D_Das_Helwig.pdf

Download (7Mb) | Preview

Abstract

Throughout the life of a transformer the effects of mechanical shocks, insulation aging, thermal processes and short circuit forces will cause deformations in the winding. This deformation can lead to vibration in the transformer and mechanical fatigue of the solid insulation. Defects which form in a transformers structure can cause faults such as partial discharge, hot spots and arcing. These faults generate combustible gases which can be analysed for condition assessment of the transformer.

The development of a suitable and cost effective vibration measurement system forms a key part of this research project. A monitoring system is developed for real-time vibration analysis. An embedded capacitive accelerometer is used in conjunction with an Arduino microcontroller to record vibrations. The sensor platform is designed to communicate wirelessly via XBee radios to a terminal computer. A software program and user interface is designed as a tool for analysis.

The outcomes and benefits of these works are primarily based on determining the condition of transformer insulation through measurements of vibration. Following a working measurement system, suitable transformer sites are monitored. Spectral analysis is performed in the frequency domain to determine a correlation with gas analysis results. The validity of vibroacoustic measurement as a predictive maintenance tool is subsequently evaluated.

Six transformers are chosen for vibration monitoring with analysis of the vibration signatures correlated to the dissolved gas analysis reports at each site. The vibration signatures at each location are analysed using the Short Time Fourier Transform and frequency peaks compared for the different sites. It was noted that sensor location does not have a large impact on vibration magnitudes and identifying the frequency components present in the signal. However, from the signatures obtained there is not enough variation in magnitude or frequency components to suggest that this method can identify the type of fault present.


Statistics for USQ ePrint 31486
Statistics for this ePrint Item
Item Type: USQ Project
Item Status: Live Archive
Additional Information: Bachelor of Engineering (Honours) Major Power Engineering project
Faculty / Department / School: Current - Faculty of Health, Engineering and Sciences - School of Mechanical and Electrical Engineering
Supervisors: Helwig, Andreas; Das, Narottam
Date Deposited: 24 Jul 2017 01:11
Last Modified: 24 Jul 2017 01:11
Uncontrolled Keywords: transformers; vibroacoustic; combustible gases; vibration measurement system; spectral analysis
Fields of Research : 09 Engineering > 0906 Electrical and Electronic Engineering > 090602 Control Systems, Robotics and Automation
URI: http://eprints.usq.edu.au/id/eprint/31486

Actions (login required)

View Item Archive Repository Staff Only