Using existing metering infrastructure, Electrotek Concepts PQView® power monitoring software application and a distribution model application such as Eaton CYMDIST® makes it possible to determine location of distribution faults.
In following examples, Hydro Ottawa has used this method to prevent future outages. In both examples, Hydro Ottawa is using their fleet of Schneider ION® revenue meters and multi-tasking those to also provide power quality data including waveform capture of events.
This data is imported by their PQView power monitoring software application. They also import CYMEDIST® feeder models into PQView and the FaultPoint® software module calculates “Reactance To Fault” (XTF) and determines where the predicted faults occur on the feeder. These findings can be plotted on Google Earth or other KML file viewers.
Example One: Multiple Faults – Same Location
In the first example a “C” phase single-phase fault occurs on three occasions within 5 months resulting in momentary outages. An example of the event voltage and current waveform is shown below as Figure
Figure 1: Typical event signature for three single phase faults.
The first event occurred on July 24th 2015 and PQView estimated the location of the fault. All predicted fault locations in the area are shown in Figure 2. Red squares with black circles are predicted locations and red squares with black stars are actual locations. The red lines are used to show predicted location +/- 5%. A crew was dispatched to inspect the area and nothing was found. On October 6th 2015, a similar event occurred and same location was predicted. A crew responded and requested that more tree trimming be completed (shown in the yellow box) because of their close proximity to the troubled feeder. On December 3rd 2015 a similar event again occurred and a crew was advised to take a close look at components to determine the cause for these events. Closer inspection from an aerial device found a split in a cable termination on a “C” phase cable. We believe this damage was the cause for the three events and no further events were captured at this location after the cable was repaired.
Figure 2: Example 1 Predicted and Actual fault locations
Example Same Fault – Multiple Feeders
The second example was a simultaneous fault on two feeders in the same area that evolved from a two-phase to three-phase fault therefore we assumed it was on the three-phase trunk line ruling out any faults on underground single phase cables. Figure 3 shows what the waveform event looks like on both feeders. PQView analyzed both events and predicted fault locations were exported to Google Earth.
Figure 4 shows the fault on feeder 1 source from the north (top) and 1 section of overhead 3-phase line was identified. Figure 5 shows the fault on feeder 2 source is from the west (left) and 2 locations are predicted with similar impedance values.
Figure 3: Event signature for multi-phase fault on multiple feeders
Figure 4: Example Two, Feeder 1 Predicted and Actual fault locations
Figure 5: Example Two, Feeder 2 Predicted and Actual fault locations
Crews patrolled the overhead sections and nothing was found. The weather conditions were good and there was no sign of animal contact or vehicle accident. Next day after reviewing the data, a crew was dispatched to inspect two air insulated pad-mounted switchgear in the vicinity. They found one of the switchgear damaged (identified as “Actual Fault Location” on Fig. 4 & 5) and believed another fault was imminent. The switchgear was taken out of service and repaired without any other problems or unplanned outages.
In both examples Hydro Ottawa prevented further outages for their customers and prevented further damage and strain to their cables and breakers
CPS Current Power Services (2016) Ltd is the Canadian distributer for PQView in Canada and I would be happy to provide more details if you wish to contact me at email@example.com