What outage attributes affect the frequencies of forced transformer outages in Eskom regions? Evidence from 2007–2016 data

Abstract

In this paper, the authors present a novel approach to the study of the frequency of forced outages in various operating units, or regions of Eskom, using statistical techniques. The forced outage data are summarized in frequencies and rates of forced outages for various levels, providing insight into variations among the levels of variables, and helping to identify levels that could be targeted in forced outage reduction strategies. The chi-squared test was performed to identify variables that had associations with the operating units (OUs), isolating variables that could be focused on to reduce frequencies in OUs. Where dependence was determined, the degree of association was calculated, enabling the ranking of the dependence with OUs in terms of which ones, if targeted, will have the highest impact on the frequency of forced outages in OUs. The paper contributes to the discourse on outages of equipment and provides insight into historical forced outage data. Furthermore, the paper provides a practical approach to studying the frequency of forced outages in regions of a utility. Finally, the insight gained from the study can enhance decision making related to the asset and performance management of power system equipment in general, and subtransmission transformers in particular.

Keywords

• Chi-square test
• Forced outage
• Transformer