Optimisation Of Online Condition Monitoring To Predict Power Transformer Failures In Distribution Networks

Abstract

The stability of any distribution network is very much dependant on the performance of the power transformers. These transformers need to be monitored to ensure that any unwanted conditions, including high risk defects, are detected. The previous and current transformer failures in the Northern Cape Operating Unit show that the health status in the Plant Health Indicator of the transformers was on category A. Category A status means the transformers were healthy. The root causes of failure were that only some of the Dissolved Gas Analysis parameters were monitored. The transformers were shown to be healthy by the Plant Health Indicator at the time of the failure. Parameters such as cooling, bushings, tap changers, operating temperatures, ageing and loss of life were not monitored. The majority of transformer failures recorded were affected by the parameters mentioned above, of which some are not monitored in the Plant Health Indicator. This research study discusses the application of integrated online condition monitoring for power transformers as one of the techniques to assess the condition of the power transformer during operation. The integrated online condition monitoring allows for transformer condition assessment without having to switch off the transformer, thereby minimising power supply interruptions. The integrated online condition monitoring system is unique technology that monitors the condition of various parameters of the power transformer such as the cooling fans, on-load tap changer, bushings, transformer operating temperatures, dissolved gas analysis, transformer ageing and loss of life. The research design is quantitative with data analysis retrieved directly from the online condition monitoring server. The results from the bushings, cooling fans, tap changer and temperatures form part of the analysis. Data analysis lead to a discussion of these parameters, which highlights the problems currently experienced with transformer failures in the Northern Cape Operating Unit. The research study revealed that, although Eskom is having the best maintenance practices in power transformers, there is a need to improve and save costs, as the current maintenance strategies of power transformers are expensive. The results of the study should be regarded as exploratory, and provide directions to researchers for further, more in-depth studies in this field. The cost analysis was performed over a period of ten years, as compared to the current costs of maintenance/monitoring and asset replacement strategies of power transformers. The total saving per year was a great benefit for tasks such as inspections/maintenance/monitoring and asset replacement. This was validated by the results from the proposed transformer online monitoring technology that can predict the defects which may result in catastrophic transformer failures.