Skip to Main Content
The first article in this issue is "Partial Discharge Diagnostics: From Apparatus Monitoring to Smart Grid Assessment" by Giancarlo Montanari and Andrea Cavallini, University of Bologna, Italy. It is the third in a series of reviews to be published in the Magazine to mark the 50th anniversary of DEIS. In it common knowledge of partial discharge mechanisms and measurements is reviewed, in order to explain why application of this powerful diagnostic technology has not been as widespread as might be expected. Condition-based maintenance of electrical apparatus based on PD detection has been moving toward on-line measurements for at least the last 20 years. So why is there still considerable activity in off-line PD testing today, probably even more than in on-line testing? The answer is straightforward: offline measurements offer the possibility of reduced noise, whereas on-line measurements may be severely affected by noise and disturbance from electrical apparatus other than that under test. Noise and disturbance rejection is essential for greater acceptability of on-line PD detection. The authors discuss some of the methods which have been proposed recently for noise rejection, based on time-of-flight techniques or on decomposition of the acquired PD pulse waveforms. Identification of the type of defect responsible for a given PD pattern is another important task, at present usually carried out by "experts." However, this approach is incompatible with on-line, conditionbased- maintenance practices, because of the very large data stream generated by on-line monitoring, and the likely scarcity of experts. Thus efforts are being made to develop artificial intelligence tools, such as fuzzy logic or neural networks.