Researchers are developing a wireless sensor network (WSN) designed to spot faults in electricity sub-stations that can lead to power cuts. The EPSRC-funded team will develop a WSN capable of sensing partial discharge (PD) in electricity sub-stations, a situation that occurs when the insulation of cables and other power equipment becomes old or damaged. Left unchecked, partial discharge can lead to dangerous and destructive faults including explosions and power cuts. Designed to be monitored centrally, the new WSN will allow operators to replace planned maintenance with condition-based maintenance.
Ian Glover, the new Professor of Radio Science and Wireless Systems Engineering at Huddersfield University told The Engineer via email that the traditional approach to PD detection using free-standing radio receivers has been to measure the difference in time-of-flight from the PD source to a set of spatially separated receivers.
‘The difference in the times-of-flight are found by cross-correlating the noise-like time waveforms arriving at the different receivers with each other,’ he said. ‘The difference in the times-of-flight for a pair of receivers defines a locus of points on which the source of PD could lie. Multiple loci, resulting from multiple pairs of receivers, intersect which gives the location of the source.’ The 4.5 year project, which has received £670,000 in funding, aims to develop a system that relies principally on measurement of PD signal amplitude and does not rely on time measurements. One challenge, said Prof Glover, will be to make the sensors sensitive enough to detect PD at a useful range without requiring sophisticated signal processing, such as the cross-correlation used in the time-of-flight approach. He said, ‘Such signal processing is power hungry and these sensors will probably need to be powered using energy harvesting technologies – solar cells, vibration, stray electric and magnetic fields, for example – if they are not to require expensive maintenance.’
Another challenge, he said, is that the attenuation [loss] of the PD signal in propagating from source to receiver may vary significantly, even for paths of the same length due to the complex propagation environment of the substation.
‘This means that the location of the PD source is almost certainly not possible by simply inverting a path loss law since the path loss law will be unknown,’ said Prof Glover. ‘It may be that we have to ‘calibrate’ our sensors using an emulated PD signal. This itself will require power and may further challenge the energy harvesting solution to maintenance avoidance.’
More info here.