Analysis of on-line power cable signals

For many years identification of incipient Partial Discharge (PD) faults in power cables has been made through off-line investigation techniques. The periodic monitoring of power cables resulted in unexpected failures, with consequent financial penalties for the utilities. More recently, in an effort to allow continuous asset management of the medium voltage cable network to be carried out, on-line monitoring systems are being installed with the aim of reducing unexpected failures. This paper presents work on the analysis and handling of data acquired from an on-line system. A short review of on-line vs. off-line cable PD monitoring will be presented, in terms of their respective advantages and disadvantages. The authorspsila experience of applying Wavelet-based denoising techniques to extract PD data from external noise interference will be presented. Analysis of PD activity and noise interference, with respect to time-of-day, will offer insight into the challenges relating to these systems. Finally, a means of handling the vast amounts of data and of acquiring knowledge from on-line condition monitoring data will be discussed.