The use of artificial neural networks for condition monitoring of electrical power transformers

Abstract

Condition monitoring of electrical plant represents an area of great interest to both manufacturing and utility companies within the electricity supply industry. De-regulation and privatisation entail that utilities must operate their systems in an optimal fashion and one of the technologies which can facilitate this is condition monitoring. Condition monitoring has a number of important benefits: unexpected failures can be avoided through the possession of quality information relating to the on-line condition of the plant and the consequent ability to identify faults or problems while still in the incipient phases of development; maintenance programmes can be condition based rather than periodically based; the plant may be utilised more optimally through the use of information relating to the plant's real-time condition and/or performance – for example, the plant may be driven temporarily beyond its stated capacity if it is known that this will not cause any short-term problems. This paper will cover the generic capabilities of artificial neural networks, in both estimation and classification mode, for condition monitoring applications, using examples based around work that the authors have carried out with respect to the monitoring of a power transformer.