As a multi-utility in the United Kingdom with business activities in the electricity, water, gas and telecommunications industries, ScottishPower is keenly aware of the increasing standards of service demanded by today's customer. ScottishPower investigated the problems caused by voltage sags, transients and harmonic distortion on its system by consulting with customer groups to evaluate their needs and determine their perception of quality power supply. Customer concerns range from the impact of short interruptions of supply (due to rural reclosers) to voltage sags on sensitive production process equipment.

ScottishPower implemented an extensive monitoring program to help identify quality of supply characteristics at a number of sites on its system. The utility has used the collected data to identify where power system performance can be improved and where customers can improve the resilience of their processes to voltage sags. The remote monitoring systems focus on power quality monitoring, protection performance monitoring and fault location (impedance to fault and a traveling wave location system).

Power Quality Monitoring Access to data on voltage sags is paramount in determining the severity of a problem and in assessing required improvements. Knowledge of sag depth and duration is essential to understanding a customer's process resilience and plant ride through capability. Data from ScottishPower's 11-kV distribution system is captured by disturbance monitors (type DL8000 as produced by Leading Edge Research of Northern Ireland) installed at more than 100 primary substations in Scotland and in the northwest England and north Wales areas of Manweb, a ScottishPower subsidiary (Fig. 1).

Voltage measurements are captured from 11-kV potential transformer secondary supplies. The DL8000 logs the triggered data, and trigger levels are independently set for each monitor. All monitors are accessed directly via modem from ScottishPower's Power Systems head office in Glasgow, Scotland. An auto-polling process gathers disturbance data overnight from all recorders. The data is server based and can be accessed via desktop PCs by Power Systems staff. Ad hoc access to individual monitors can also be gained at any time.

Power quality profiles are site specific and, although average profiles can be assumed, many customers require detailed information to provide the basis for power quality investment decisions. By comparing recorded voltage depressions with the known performance of the customer's plant, customers are able to identify the vulnerable processes and prepare an action plan to minimize plant disruption. The data collected can be used to compare system performance after the investment has been made to upgrade protection systems.

Figure 2 shows voltage sag data collated before and after modifications were made to radial distribution protection schemes to provide faster clearance times. This exercise, in conjunction with a customer's own system improvements, provided an improved process resilience to voltage sags.

Monitoring The units at primary substations (33 kV/11 kV) supplying overhead line networks capture both voltage and current data as a means of monitoring the performance of overhead line protection systems. ScottishPower has performed this monitoring in conjunction with distribution protection upgrades aimed at eliminating outages due to transient faults. The recorders are set to trigger on rate-of-rise of current, which enables detection of downstream protection devices operating on low current faults. Coupled with digital inputs from source feeder protection starters, the recorders can determine which downstream device operated without relying on SCADA at remote breakers. Figure 3 shows how the protection upgrade greatly decreased transient fault lock-outs following the correction of initial problems detected by the monitoring system.

Impedance to Fault By transferring the distribution system data into another software package, ScottishPower can generate an impedance-to-fault value as a means of predicting the fault location. This technique is being developed for phase-to-phase faults with successful results. For transient faults it becomes possible to locate and rectify defects before further customer interruptions and before a permanent fault develops.

This proactive approach has proved beneficial in providing transient fault locations so that remedial action can be taken to prevent repetition of events. For example, the system identified a repeated flashing phase-to-phase fault, and the consistent impedance-to-fault measurement provided a location at 3.9 km (2.4 mile) on a 20-km (12.4 mile) circuit. Close inspection by local repair crews identified fresh arcing marks at a mid-span location of 150 m (492 ft) from the identified location. Remedial action eliminated the problem and prevented further customer interruptions.

Traveling Wave System (TWS) The value of pre-locating faults is more pronounced on transmission lines with typical line lengths of 140 km (87 miles). ScottishPower uses a traveling wave system, developed in conjunction with Hathaway Ltd., Hoddesdon, United Kingdom, for transmission line fault location. The system is aimed at improving quality of supply by reducing the incidence of faults on the network. Transient faults can be identified and repair crews dispatched to rectify faults before further disruption is caused. The accuracy of the fault location system is typically better than %200 m (656 ft), which allows for rapid fault identification. CT secondary circuits detect current transients using inductive sensors with a sample rate of 1.25 MHz. When a fault occurs on a power line, a traveling wave transient emanates in both directions from the fault and can be detected at both ends of the line. The propagation depends on line construction but is typically 200 m/micro sec.

Two methods of fault location are available: Single Ended - Lets say the transient wave will arrive at end A and is reflected back to the fault. The fault arc, which is present for up to 100 msec, will reflect the transient back to end A. The time of flight between these two events, detected at end A, determines the distance to fault from end A. Double Ended - The transient impulse is received at end A and at end B at different times. The time difference is a measure of the distance to the fault from each end. To achieve accurate time tagging of the event, the detectors are synchronized to a GPS clock that provides a system accuracy of 1 micro sec, equivalent to %200 m fault location distance. By minimizing trace interpretation, the double-ended scheme is the preferred location method. Figure 5 shows an example of the results produced by this method.

These initiatives keep ScottishPower at the forefront of technological advances and enable cost-effective techniques to provide quality power supply to its customers.

Alec Burden is the quality and performance manager of the Power Systems Division of ScottishPower. Burden has worked in all aspects of electrical distribution systems including cable fault location, protection, metering, operational safety, power quality, live-line working, control room systems and research and development. He is a chartered engineer and a member of the Institution of Electrical Engineers.

Alan MacGregor joined South of Scotland Electricity Board (ScottishPower pre-privatization) after graduating with the BSc electrical and electronic engineering from Heirot-Watt University, Edinburgh in 1977. He has experience in various aspects of distribution including construction, maintenance, protection systems and cable fault location. MacGregor is a system performance engineer in the Power Systems Group, headquartered in Glasgow. He is a chartered engineer and a member of the Institution of Electrical Engineers.