POWER TRANSFORMER END-OF-LIFE MODELLING: LINKING STATISTICAL APPROACH WITH PHYSICAL AGEING PROCESS

Replacing aged transformers requires intensive capital investment and is expensive for utilities in developed countries. A transformer end-of-life model expressed as population failure hazard curve against age is needed in order to predict the number of failure in the future and suggest the long-term capital investment. In this paper the statistical analysis on historical failure data is used to produce the failure hazard in the normal operating stage dominated by the random failure mechanism, while a physical ageing model is built based on scrapping transformer information. By linking the ageing failure mechanism with statistical approach, the population hazard curve against age is produced which can be used by operators to identify the transformer failure number a system can tolerate while maintaining a secured system operation.